1.A.3.c - Railways

Last updated on 20 Mar 2018 07:37 (cf. Authors)

Short description

In category 1.A.3.c - Railways, emissions from fuel combustion in German railways and from the related abrasion and wear of contact line, braking systems and tyres on rails are reported.

NFR-Code Name of Category Method AD EF Key Category 1
1.A.3.c Railways T1, T2 NS, M CS, D, M L: TSP, PM2.5 | L & T: PM10

Germany's railway sector is undergoing a long-term modernisation process, aimed at making electricity the main energy source for rail transports. Use of electricity, instead of diesel fuel, to power locomotives has been continually increased, and electricity now provides 80% of all railway traction power. Railways' power stations for generation of traction current are allocated to the stationary component of the energy sector (1.A.1.a) and are not included in the further description that follows here. In energy input for trains of German railways, diesel fuel is the only energy source that plays a significant role apart from electric power.

Method

Activity Data

Basically, total inland deliveries of diesel oil are available from the National Energy Balances (NEBs) (AGEB, 2017) [1]. This data is based upon sales data of the Association of the German Petroleum Industry (MWV) [2]. As a recent revision of MWV data on diesel oil sales for the years 2005 to 2009 has not yet been adopted to the respective NEBs, this original MWV data has been used for this five years.

Data on the consumption of biodiesel in railways is provided in the NEBs as well, from 2004 onward. But as the NEBs do not provide a solid time series regarding most recent years, the data used for the inventory is estimated based on the prescribed shares of biodiesel to be added to diesel oil.

Small quantities of solid fuels are used for historical steam engines vehicles operated mostly for tourism and exhibition purposes. Official fuel delivery data are available for lignite, through 2002, and for hard coal, through 2000, from the NEBs. In order to complete these time series, a study was carried out in 2012 by Probst & Consorten. [3] During this study, questionaires were provided to any known operator of historical steam engines in Germany. Here, due to limited data archiving, nearly complete data could only be gained for years as of 2005. For earlier years, in order to achieve a solid time series, conservative gap filling was applied.

Table 1: Overview of activity-data sources for domestic fuel sales to railway operators
Activity data source / quality of activity data
combustion of:
Diesel oil 1990-2004: NEB lines 74 and 61: 'Schienenverkehr' / 2005-2009: MWV annual report, table: 'Sektoraler Verbrauch von Dieselkraftstoff' / from 2010: NEB line 61
Biodiesel calculated from official blending rates
Hard coal 1990-1994: NEB lines 74; 1995-2004: interpolated data; from 2005: original data from studies; 2016: forward extrapolation
Hard coal coke 1990-1997: NEB lines 74 and 61; 1998-2004: interpolated data; from 2005: original data from studies; 2016: forward extrapolation
Raw lignite from 1990: NEB lines 74 and 61
Lignite briquettes from 1990: NEB lines 74 and 61
abrasion and wear of contact line, braking systems and tyres on rails:
transport performance data in Mio ptkm (performance-ton-kilometers) derived from the TREMOD model
Table 2: Annual fuel consumption in German railways, in terajoule
1990 1995 2000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Diesel Oil 38,458 31,054 25,410 18,142 17,101 16,730 16,389 14,336 14,626 14,730 13,514 13,771 12,283 13,321 12,381
Biodiesel 0 0 0 397 498 747 810 987 949 966 882 798 745 720 650
Liquids TOTAL 38,458 31,054 25,410 18,539 17,599 17,477 17,199 15,324 15,575 15,696 14,396 14,569 13,028 14,041 13,031
Lignite Briquettes 0 0 431 0 0 0 0 0 0 0 0 0 0 0 0
Raw Lignite 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Hard Coal 576 250 250 255 262 255 300 321 314 345 357 352 341 339 340
Hard Coal Coke 0 86 1 1 1 1 1 1 1 1 1 1 1 1 1
Solids TOTAL 576 336 682 256 263 256 301 322 315 346 357 353 342 340 341

The use of other fuels – such as vegetable oils or gas – in private narrow-gauge railway vehicles has not been included to date and may still be considered negligible.

Table 3: Annual transport performance, in Mio ptkm (performance-ton-kilometers)
1990 1995 2000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Electric Traction 36,151 33,785 36,163 35,661 37,571 37,955 37,777 33,658 34,455 34,270 35,008 33,530 33,124 32,339 29,580
Diesel Traction 9,881 5,880 3,724 2,654 2,557 2,266 3,175 2,773 2,670 2,740 2,679 2,377 2,373 2,140 2,148
over-all 46,033 39,666 39,887 38,315 40,128 40,222 40,952 36,431 37,125 37,010 37,688 35,907 35,497 34,478 31,728

Regarding particulate-matter and heavy-metal emissions from abrasion and wear of contact line, braking systems, tyres on rails, annual transport performances of railway vehicles with electrical and Diesel traction derived from (ifeu, 2017a) [4] are applied as activity data.

Emission factors

The (implied) emission factors used here for estimating emissions from diesel fuel combustion of very different quality:
For main pollutants, CO and PM, annual tier2 IEF computed within the TREMOD model are used, representing the development of German railway fleet, fuel quality and mitigation technologies [4].
On the other hand, constant default values from (EMEP/EEA, 2016) [5] are used for all reported PAHs and heavy metals and from (Rentz et al., 2008) [6] regarding PCDD/F.
As no emission factors are available for HCB and PCBs, no such emissions have been calculated yet.

Regarding emissions from solid fuels used in historic steam engines, all emission factors displayed below have been adopted from small-scale stationary combustion.

Furthermore, regarding emissions from abrasion and wear, emission factors are calculated from PM10 emission estimates directly provided by the German railroad company Deutsche Bahn AG.
As these original emissions are only available as of 2013, implied EF(PM10) were calculated from the emission estimates extrapolated backwards from 2013 to 1990 and the transport performance data available from TREMOD.
Regarding PM2.5, and TSP, due to leck of better information, a fractional distribution of 0.5 : 1 : 1 (PM2.5 : PM10 : TSP) is assumed for now.
Emission factors for emssions of copper, nickel and chrome are calculated via typical shares of the named metals in the contact line (copper) and in the braking systems (Ni and Cr). Other heavy metals contained in alloys used for the contact line (silver, magnesium, tin) are not taken into account here. Furthermore, emissions from other wear parts (e.g. the current collector) are not estimated. However, these components are not supposed to contain any of the nine heavy metals to be reported here (current collectors are made of aluminium alloys and coal).

Table 3: Annual country-specific emission factors from TREMOD, in kg/TJ
1990 1995 2000 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Diesel fuels 1
NH3 0.54 0.54 0.54 0.54 0.54 0.54 0.54 0.54 0.54 0.54 0.54 0.54 0.54 0.54 0.54
NMVOC 109 100 90 65 62 58 57 54 56 59 47 44 43 41 41
NOx 1,170 1,207 1,225 1,111 1,058 1,029 1,011 1,001 986 1,010 921 882 890 846 830
SOx 191 61 14 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32 0.32
PM 2 44 44 37 23 22 21 20 18 19 20 17 15 15 15 15
BC 3 29 28 24 15 15 14 13 12 12 13 11 10 10 9 9
CO 29 28 24 15 15 14 13 12 12 13 11 10 10 9 9
Hard coal
NH3 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00
NMVOC 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15
NOx 120 120 120 120 120 120 120 120 120 120 120 120 120 120 120
SOx 650 650 650 650 650 650 650 650 650 650 650 650 650 650 650
PM2.5 222 222 222 222 222 222 222 222 222 222 222 222 222 222 222
PM10 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250
TSP 278 278 278 278 278 278 278 278 278 278 278 278 278 278 278
BC 3 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14
CO 500 500 500 500 500 500 500 500 500 500 500 500 500 500 500
Hard coal coke
NH3 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00 4.00
NMVOC 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50
NOx 120 120 120 120 120 120 120 120 120 120 120 120 120 120 120
SOx 500 500 500 500 500 500 500 500 500 500 500 500 500 500 500
PM 2 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15
BC 3 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96 0.96
CO 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000

1 due to lack of better information: similar EF are applied for fossil diesel oil and biodiesel
2 EF(PM2.5) also applied for PM10 and TSP (assumption: >99% of TSP consists of PM2.5)
3 EFs calculated via f-BCs as provided in [5]: diesel fuels: 0.56 (Chapter: 1.A.3.c - Railways, Appendix A: tier1), solid fuels: 0.064 (Chapter: 1.A.4 - Small Combustion: Residential combustion (1.A.4.b): Table 3-3, Zhang et al., 2012)

Table 4: Country-specific emission factors for abrasive emissions, in g/km
PM2.5 PM10 TSP BC Pb Cd Hg As Cr Cu Ni Se Zn
Contact line 1 0.00016 0.00032 0.00032 NA NA NA NA NA NA 0.00033 NA NA NA
Tyres on rails 2 0.009 0.018 0.018 NA NA
Braking system 3 0.004 0.008 0.008 NA NA NA NA NA 0.00008 NA 0.00016 NA NA
Current collector 4 NE NE NE NE NA

1 assumption: 100 per cent copper
2 assumption: 100 per cent steel
3 assumption: steel alloy containing Chromium and Nickel
4 typically: aluminium alloy + coal contacts; no particulate matter emissions calculated yet

Table 5: Tier1 emission factors for heavy-metal and POP exhaust emissions
Pb Cd Hg As Cr Cu Ni Se Zn B[a]P B[b]F B[k]F I[…]p PAH 1-4 HCB PCBs PCDD/F
[g/TJ] [mg/TJ] [µg/TJ]
Diesel fuels 1.21 2 0.23 1 0.12 1 0.002 2 1.16 2 39.57 2 1.63 2 0.23 2 23.28 2 698 2 1,164 2 801 1 184 1 2,847 3 NE NE 2.09
Lignite Briquettes NE 34,500 NE NE NE 90,000 NE NE 29.80
Raw Lignite NE NE NE NE NE
Hard Coal NE NE NE NE NE
Hard Coal Coke NE NE NE NE NE

1 tier1 default from [5], chapter: 1.A.3.b i-iv - Road transport: exhaust emissions: tier1 value for diesel vehicles
2 tier1 default from [5], chapter: 1.A.3.c - Railways
3 sum of tier1 default value applied for B[a]P, B[b]F, B[k]F, and I[…]P

Discussion of emission trends

NFR 1.A.3.c is no key source.

Basically, for all unregulated pollutants, emission trends directly follow the trend in over-all fuel consumption.

Here, as emission factors for solid fuels tend to be much higher than those for diesel oil, emission trends are disproportionately effected by the amount of solid fuels used.
Therefore, for the main pollutants, carbon monoxide, particulate matter and PAHs, emission trends show remarkable jumps especially after 1995 that result from the significantly higher amounts of solid fuels used.

For all fractions of particulate matter, the majority of emissions generally result from abrasion and wear and the combustion of diesel fuels. Additional jumps in the over-all trend result from the use of lignite briquettes (1996-2001). Here, as the EF(BC) for fuel combustion are estimated via fractions provided in [5], black carbon emissions follow the corresponding emissions of PM2.5.

Due to fuel-sulphur legislation, the trend of sulphur dioxide emissions follows not only the trend in fuel consumption but also reflects the impact of regulated fuel-qualities.
For the years as of 2005, sulphur emissions from diesel combustion have decreased so strongly, that the over-all trend shows a slight increase again due to the now dominating contribution of sulphur from the use of solid fuels.

Regarding heavy metals, emissions from combustion of diesel oil and from abrasion and wear are estimated from tier1 default emission factors.
Therefore, the emission trends reflect the development of diesel use and - for copper, chromium and nickel emissions resulting from the abrasion & wear of contact line and braking systems - the annual transport performance (see description of activity data above).

Recalculations

Activity data

Given the revised NEB 2015, both the activity data fo diesel oil and the annual amounts of blended biodiesel were revised accordingly.

Table 6: Revised 2015 fuel consumption, in terajoule
Diesel Oil Biodiesel
Submission 2018 13,321 720
Submission 2017 11,989 652
absolute change 1,332 68
relative change 11.11% 10.45%

Emission factors

Due to the routine revision of the TREMOD model [3], tier2 emission factors changed for recent years.

Table 7: Revised (I)EFs 2015 for emissions from diesel fuels, in kg/TJ
NOx NMVOC CO PM BC
Submission 2018 845.79 41.14 99.94 14.58 9.48
Submission 2017 846.46 41.18 100.11 14.59 9.49
absolute change -0.67 -0.03 -0.17 -0.02 -0.01
relative change -0.1% -0.1% -0.2% -0.1% -0.1%

Furthermore, emission factors for the estimation of emissions of particulate matter and heavy metals from abrasion & wear of contact line and braking systems are implemented for the first time.

Table 8: Newly implemented emission factors, in g/km
PM101 PM2.52 TSP3 Cr4 Ni5 Cu6
wear of braking system 0.0060 0.0030 0.0060 0.00006 0.00012 NA
wear of contact line 0.0003 0.0001 0.0003 NA NA 0.0003

1 EF(PM10) estimated from original PM10 emissions provided from the German railroad company Deutsche Bahn AG and transport performance data available from TREMOD-
2 EF(PM2.5) estimated as a fraction of 50% of the EF(PM10)
3 EF(TSP) similar to the EF(PM10)
4 EF(Cr) estimated via an assumed chromium content in steel used for braking systems of 1 per cent
5 EF(Ni) estimated via an assumed nickel content in steel used for braking systems of 2 per cent
6 EF(Cu) estimated via an assumed copper content in the contact line of 100 per cent

For specific information on recalculated emssion estimates reported for Base Year and 2015, please see the pollutant specific recalculation tables following chapter 8.1 - Recalculations.

Uncertainties

Uncertainty estimates for activity data of mobile sources derive from research project FKZ 360 16 023: "Ermittlung der Unsicherheiten der mit den Modellen TREMOD und TREMOD-MM berechneten Luftschadstoffemissionen des landgebundenen Verkehrs in Deutschland" by (ifeu & INFRAS, 2009) [7]. - For detailled information, please refer to the project's final report here (German version only!)

Planned improvements

Besides the scheduled routine revision of TREMOD, no further improvements are planned for the next annual submission.

FAQs

Why are similar EF applied for estimating exhaust heavy metal emissions from both fossil and biofuels?

The EF provided in [5] represent summatory values for (i) the fuel's and (ii) the lubricant's heavy-metal content as well as (iii) engine wear. Here, there might be no heavy metals contained in the biofuels. But since the specific shares of (i), (ii) and (iii) cannot be separated, and since the contributions of lubricant and engine wear might be dominant, the same emission factors are applied to biodiesel.


Bibliography
1. AGEB, 2017: Working Group on Energy Balances (Arbeitsgemeinschaft Energiebilanzen (Hrsg.), AGEB): Energiebilanz für die Bundesrepublik Deutschland; URL: http://www.ag-energiebilanzen.de/7-0-Bilanzen-1990-2015.html, Köln & Berlin, 2017.
2. MWV, 2017: Association of the German Petroleum Industry (Mineralölwirtschaftsverband, MWV): Annual Report 2017, page 65, Table 'Sektoraler Verbrauch von Dieselkraftstoff 2012-2015'; URL: https://www.mwv.de/wp-content/uploads/2016/06/170629_Mineraloelwirtschaftsverband_Jahresbericht-2017.pdf, Berlin, 2017.
3. Probst & Consorten, 2012: Recherche des jährlichen Kohleeinsatzes in historischen Schienenfahrzeugen seit 1990 ('Inquiry of annual coal use in historic railway vehicles since 1990') - Study carried out for UBA; FKZ 363 01 392; not yet published
4. ifeu, 2017a: Knörr, W. et al.; ifeu Institute for Energy and Environmental Research (Institut für Energie- und Umweltforschung Heidelberg gGmbH, ifeu): Fortschreibung des Daten- und Rechenmodells: Energieverbrauch und Schadstoffemissionen des motorisierten Verkehrs in Deutschland 1960-2030, sowie TREMOD, im Auftrag des Umweltbundesamtes, Heidelberg, 2017.
5. EMEP/EEA, 2016: EMEP/EEA air pollutant emission inventory guidebook 2016, Copenhagen, 2016.
6. Rentz et al., 2008: Nationaler Durchführungsplan unter dem Stockholmer Abkommen zu persistenten organischen Schadstoffen (POPs), im Auftrag des Umweltbundesamtes, FKZ 205 67 444, UBA Texte | 01/2008, January 2008 - URL: http://www.umweltbundesamt.de/en/publikationen/nationaler-durchfuehrungsplan-unter-stockholmer
7. ifeu & INFRAS, 2009: IFEU – Institut für Energie- und Umweltforschung Heidelberg gGmbH und INFRAS Zürich: Ermittlung der Unsicherheiten der mit den Modellen TREMOD und TREMOD-MM berechneten Luftschadstoffemissionen des landgebundenen Verkehrs in Deutschland, FKZ 360 16 023, Heidelberg & Zürich.
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License