Moped emissions in urban areas

Abstract

Road-traffic emissions are significant contributors to air quality degradation. In particular in urban areas traffic emissions affect the human living space. Although mopeds correspond only to a rather small fraction of total road transport activity, they are an important contributor to HC emissions in specific areas. In particular, many inhabitants of metropolitan areas in the Netherlands complain about driving behaviour, noise and stench of mopeds. At the same time mopeds become increasingly popular, in some cities the number of mopeds almost doubled over the last decade. The ubiquitous smell of mopeds in busy inner cities does raise concern over possible negative health effects.

Full text

1
Moped emissions in urban areas
P. van Mensch1*, A.R.A. Eijk1, N.E. Ligterink1,
1 Research Group Sustainable Transport and Logistics, TNO, PO Box 96800, 2509 JE The Hague, the
Netherlands, presenting author: pim.vanmensch@tno.nl
Keys-words: Mopeds, ISC, emission factors, testing, urban, speed limiters, Euro 5, PEMS, on-
road, tampering, fleet composition, L-category, 2-stroke, 4-stroke
Background
Road-traffic emissions are significant contributors to air quality degradation. In particular in
urban areas traffic emissions affect the human living space. Although mopeds correspond only
to a rather small fraction of total road transport activity, they are an important contributor to HC
emissions in specific areas. In particular, many inhabitants of metropolitan areas in the
Netherlands complain about driving behaviour, noise and stench of mopeds. At the same time
mopeds become increasingly popular, in some cities the number of mopeds almost doubled
over the last decade. The ubiquitous smell of mopeds in busy inner cities does raise concern
over possible negative health effects.
Moped fleet composition in the urban area
To calculate the contribution of mopeds to local air pollution and in order to quantify the stench
problem of mopeds, TNO investigated the moped fleet composition in two large Dutch cities
(Eijk et al., 2016). Several locations were selected and license plate cameras were installed to
register license plates and vehicle speed. The license plates of all mopeds passing the selected
locations were registered 24 hours a day, 7 days a week. The information of the Dutch road
authority and the help of the moped importers was used to determine the moped fleet
composition, see Figure 1. Ten thousands of mopeds were registered during several scans in
various cities.
Figure 1: Age distribution of passing mopeds in a specific city (Eijk et al., 2016).
Dutch moped fleet composition:
In the Netherlands two different categories of mopeds are available:
a) Mopeds with a maximum construction speed of 45 km/h, used with helmet, on the road
b) Mopeds with a maximum construction speed of 25 km/h, used without helmet, on the
bicycle lanes
Both categories are sold with 2-stroke and 4 stroke engines, approved according to different
Euro standards 0, 1 and 2. For accurate calculation of the total moped emissions, these typical
moped categories have been distinguished in the moped vehicle fleet scans.

2
Maximum construction speed and number of mopeds in the Netherlands
Mopeds with a maximum construction speed of 25 km/h have become increasingly popular over
the last 5 to 10 years, this is indicated by figure 2. This indicates a shift from long-distance,
regional transport to inner-city use of mopeds. In the Netherlands, these mopeds (max. speed
of 25 km/h) do have specific advantages over the 45 km/h mopeds: it is allowed to drive these
mopeds on all bicycle lanes in cities and it is not mandatory to wear a helmet. Drivers of
mopeds (max. speed 45 km/h) sometimes are allowed on bicycle lanes in specific urban areas,
but normally drive on the roads with the cars. Furthermore, it is mandatory to wear a helmet.
Statistics Netherlands has performed a large survey on the ownership and usage of mopeds in
the Netherlands (Ewalds, et al., 2013). On 17 million people, nationally, there are over 1 million
mopeds, which has been an increasing number over decades. This is one of the highest
numbers in Europe. Furthermore, there is an increase in the annual mileage of mopeds,
approaching an average of 4000 km/annum.
Figure 2: Sales numbers of mopeds in the Netherlands (Source Bovag)
Different Euro classes
Based on the date of first registration and the European emission legislation, mopeds can be
classified according to the different Euro emission standards. The majority of the mopeds
registered by the license plate cameras are Euro 2 mopeds. A limited number of Euro 0 and
Euro 1 mopeds remain driving in the cities. Although the Euro 3 emission standard was
introduced in 2014, Euro 3 mopeds have seldom been sold and are rarely seen in the
Netherlands. This is a result of the emission legislation, after the introduction of the Euro 3
standard, manufacturers were still allowed to sell Euro 2 mopeds.
2-stroke versus 4-stroke
In general, the emission performance of 2- and 4-stroke engines is quite different. Accurate
emission calculations require detailed insight in the share of 2- and 4-stroke engines in the fleet.
This important characteristic is unfortunately not registered by the Dutch road authority (RDW).
Quite some mopeds, even of the same make and model, are sold with both 2-stroke and 4
stroke engines. Therefore, relevant moped importers and dealers were asked for help to
determine this characteristic for all registered mopeds. Up to 2010, an important share of the
moped fleet was equipped with a 2-stroke engine. From 2011, new mopeds are mainly
equipped with 4-stroke engines.
Moped modifications
Since the first moped entered the market, enthusiastic owners started to modify their mopeds to
increase maximum acceleration and/or maximum vehicle speed. The modifications range from
70861
93495 95845 94607
82562
69845
59634 63112 66054 67936
0
20000
40000
60000
80000
100000
120000
2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Sales numbers [year]
Sales numbers of mopeds (< 50 cc) in the Netherlands
Moped 45 km/h
Moped 25 km/h
All mopeds

3
removing speed limiters (especially 25 km/h mopeds) up to modified carburettors,
injection/ignition systems, exhaust systems, engine capacity and so on.
During the license plate scan period, not only license plates were registered, but vehicle speed
as well. This information is used to estimate the number of modified vehicles on the road.
A large share of the 25 km/h mopeds pass the license plate cameras driving at speeds well
above 30 km/h, as shown in the speed distribution of 25 km/h mopeds in figure 3. Although the
official construction speed is limited to 25 km/h, some of these mopeds travel up to 60 km/h,
and in some cases even up to 80 km/h (city centre). The 45 km/h mopeds show similar
behaviour, as also displayed in figure 3. A large share of these vehicles passes at speeds
ranging from 45 up to 60 km/h, in rare cases even up to 80 km/h. This shows that speed
reduction measures are removed very frequently and some mopeds are modified in a more
advanced way.
Figure 3: vehicle speed of mopeds with an official 25 and 45 km/h construction speed when
passing the license plate camera (Eijk et al., 2016).
Typical issues with mopeds in the Netherlands
Speed limiters
These aforementioned mopeds with a construction speed of 25 km/h are a typical Dutch moped
configuration. The majority of these mopeds with a maximum construction speed of 25 km/h are
technical similar to mopeds with a maximum speed of 45 km/h. However, the maximum speed
has been reduced by simplistic technical solutions. These simple measures to reduce the
maximum speed of mopeds usually raise emission levels and fuel consumption, when this
maximum speed is reached.
Commonly applied speed limiters for 4-stroke mopeds are ‘engine speed limiters’ (often applied
for vehicles which use a carburettor for the fuel supply) and ‘variomatic limiters’ (transmission
ratio limiters). Both types of speed limiters have a negative effect on fuel consumption
(Hensema et al., 2013). The engine speed limiter delays the ignition timing for the combustion in
order to restrict the engine speed. By doing so, a large part of the fuel is combusted not
delivering engine power, resulting in a high fuel consumption. By driving only just below the
limited speed, the negative effect of the engine speed limiter reduced. It is commonly known
that in real-world driving conditions, most drivers will drive those vehicles in full-throttle
operation and thus with an activated speed delimiter. The engine speed limiter especially has a
negative effect at maximum speed.
In figure 4 a graph of the fuel consumption visualizes the increase in fuel consumption for various
configurations of mopeds. Driving a moped without speed limiters at maximum allowed speed of

4
25km/h was the most fuel efficient. The delimited 25 km/h version was 3.5 times more fuel efficient
than the standard 25 km/h version (Hensema et al., 2013).
Figure 4: Visualization of the increase in fuel consumption for various configurations of mopeds
(Hensema et al., 2013).
Modified fuel nozzle to meet client expectations
Based on discussions with several dealerships of mopeds in the Netherlands, it is suspected that
many new mopeds are adjusted by dealerships before the new moped is handed over to its first
owner (Ntziachristos et al., 2017). The dealerships claim that they make this adjustment in order
to deliver a vehicle to the client that meets the client expectations: a moped with a smooth-running
engine that starts and drives well under all conditions. This adjustment often involves replacement
of the fuel nozzle by a larger one, this applies to vehicles with an engine with a carburettor.
According to the dealers, the client expectations for drivability often cannot be met without the
adjustments. However, this means that those vehicles are not compliant to the type approval
specifications anymore, though they are representative for many vehicles in-use. Still, as a result,
emissions of the vehicle that is delivered to the end-user may not comply to the emission
requirements anymore. The conformity of production (CoP) requirements ensure that vehicles are
compliant when they leave the factory. These adjustments are made after production and cannot
be detected with the current set of type approval procedures. However, this phenomenon may
result in a large number of in-use vehicles that are not compliant to the emission requirements
during their full lifetime.
But, the anti-tampering provisions of Regulation (EU) No 168/2013 are more stringent than in the
Directive 97/24/EC, which may affect the size of the issue for Euro 4 and Euro 5 compliant
mopeds. Moreover, the size of the issue might be different with introduction of vehicles which
have more advanced engines technologies to comply with the Euro 5 emission limits
(Ntziachristos et al., 2017).
Typical emissions from mopeds
Based on the moped fleet composition and appropriate emission factors, the contribution of
mopeds to the total traffic emissions and local air pollution can be calculated. This requires an
extensive and reliable set of emission factors (van Zyl et al., 2015). As mentioned before, mopeds
come in many different types. The current emission factors per type of moped with urban driving
are shown in graphs of figure 5 for multiple emission constituents. In this figure, the main bars
represent the emission factors of standard (modern) vehicles. The upper and lower stripes
represent extensively tampered mopeds and delimited mopeds respectively. The displayed
stripes for the passenger cars represent an average passenger car (2015). In the table below, a
distinction is made of the main type of mopeds with their typical emission characteristics. These
typical emission characteristics are based on the emission factors. In addition, the effect on fuel
consumption is added for some types.

5
Table 1_Typical emission characteristics per type of moped
Type of moped
Typical emission characteristics
45 km/h mopeds:
2-stroke
In general more polluting than 4-stroke, especially high HC and
particle emissions. The CO emissions are generally lower than
with 4-stroke mopeds.
45 km/h mopeds:
4-stroke
Less polluting than 2-stroke mopeds. However, CO emissions
can be excessively high.
25 km/h mopeds
Higher emissions and higher fuel consumption than 45 hm/h
mopeds.
Delimited 25 km/h
mopeds
Comparable emissions and fuel consumption as 45 km/h
mopeds.
Extensively
modified/tampered
mopeds resulting in
very high vehicle
speeds
Depending of the stage of tampering, an excessive increase of
emissions is possible, in particular when the exhaust is replaced.
Figure 5: The current emission factors per type of moped with urban driving
0
5
10
15
20
25
30
2-stroke
4-stroke
2-stroke
4-stroke
Petrol
Diesel
25 km/h 45 km/h Passenger
car
CO emissions [g/km]
0
2
4
6
8
10
12
2-stroke
4-stroke
2-stroke
4-stroke
Petrol
Diesel
25 km/h 45 km/h Passenger
car
HC emissionss [g/km]
0.0
0.1
0.2
0.3
0.4
0.5
0.6
2-stroke
4-stroke
2-stroke
4-stroke
Petrol
Diesel
25 km/h 45 km/h Passenger
car
NOx emissions [g/km]
0.00
0.02
0.04
0.06
0.08
0.10
0.12
2-stroke
4-stroke
2-stroke
4-stroke
Petrol
Diesel
25 km/h 45 km/h Passenger
car
PM10- emissions [g/km]

6
New insights from recent measurements
To refine the most important, detailed emission factors, TNO has collected new emission data in
two measurement projects. In the first project, nineteen Euro 2 mopeds, both 4-stroke and 2-
stroke mopeds, were tested on the chassis dynamometer (Ntziachristos et al., 2017). Nine of
these mopeds were in-use 4-stroke mopeds from the Netherlands. Three models with high sale
volumes were selected. Three mopeds per selected model have been located at various official
dealerships. The original anti-pollution components, i.e., the exhaust system, of these vehicles
were still present. Moreover, the condition of these vehicles was checked at the dealerships
before the vehicle was delivered to the emission test centre. However, some vehicles do not have
a complete maintenance record since not all vehicles went for maintenance to the same workshop
during their lifetime. The procedure for Type I testing as described in Annex II of Regulation (EU)
No 134/2014 is used for these tests. Every vehicle drove a precondition cycle before the actual
test. The preconditioned vehicles drove two WMTC’s with a cold start.
The graph in figure 6 shows a comparison between the emission results of measured in-use 4-
stroke mopeds and the Euro 2 emission limits, as a function on the vehicle’s mileage. The
conformity factor is determined by dividing the emission results of the test (WMTC) by the
applicable Euro 2 emission limit. Each marker shape represents a vehicle model. It should be
noted that the formal Type I test with which these vehicles should comply, was not driven. This
means that these figures do not represent formal in-service-conformity of the tested vehicles.
The experimental results of the Euro 2 mopeds in this study showed some excessively high CO
emissions for the 4-stroke mopeds, though, relatively low CO emissions were measured as well.
The HC and NOx emission performance of these 4-stroke mopeds were in general similar to the
Euro 2 emission limits. However, other Euro 2 vehicles with a 2-stroke engine which were tested
in another task of this study showed significantly higher HC emissions than the 4-stroke mopeds.
Moreover, these emissions were often higher than the previously described emission factors for
HC. Most of these other mopeds were not provided by dealerships.
The CO emission performance varies significantly per vehicle model. Moreover, the CO emission
performance can vary greatly as well between the three tested vehicles of one model. On the
contrary, the HC + NOx emission performance per vehicle model does not show significant
variations. Furthermore, the emission performance does not vary greatly between the three tested
vehicles of one model. For these tested vehicle, there is no clear relationship between the
emission performance and the period that vehicles are in service. The emissions are either high
or low, once the vehicle has driven more than 10% of its defined useful life mileage.
The vehicles are tested by using the WMTC, which is not the applicable Type I driving cycle for
the tested vehicles. However, such excessive exceedances are cannot solely be the result of
changing the driving cycle to the WMTC. It is not clear whether a part of the selected vehicles
have modified fuel nozzles, which then might cause higher emissions. Hence, it not clear if these
high emissions can be the related to the potentially made adjustments. The high emissions can
also be the result of fast degraded anti-pollution devices, as there are no durability requirements
for the tested vehicles yet, or a combination of these two issues. Alternatively, ineffective
Conformity of Production (CoP) can also be a possible cause, which is less probable. It is
important to introduce measures for in use vehicles in order to prevent such high emissions.
In general, these results from this study are in line with the previously described emission factors.
Figure 6: A comparison between emission results and the Euro 2 emission limits as a function
on the vehicle’s mileage (Ntziachristos et al., 2017).

7
During the second project, sixteen mopeds are measured. A wide range of common vehicle types
was selected, ranging from Euro 2 to Euro 3, 2-stroke to 4-stroke, new and used and from fuel
injection to carburettor. These measurements showed similar results for the emissions of 4-stroke
mopeds as described before. However, the vehicles which use electronic fuel injection, rather
than a carburettor, showed CO emissions which were significantly lower. On the contrary, the 2-
stroke mopeds showed on average higher HC and PM emissions than the current emission
factors.
Policy instruments for lowering emissions of mopeds
European legislation
In order to reduce emissions from mopeds, the European Commission introduced several
stages of emissions control regulation since 1997. Directive 97/24/EC defined emissions limits
up to the Euro 3 step. Since 2017, the Euro 4 step is implemented with more stringent emission
limits. Moreover, the Euro 4 step has a comprehensive package of environmental tests,
including durability requirements, a fuel permeability test and a crankcase emission test. The
intention has been not only to make sure that new vehicle models become increasingly clean
when placed to the market but also that they remain so throughout their useful life. In 2020 the
Euro 5 step will be implemented. In this step, the emission limits are comparable with the
emission limits for Euro 6 passenger cars. Moreover, the WMTC, a more dynamic test cycle, is
introduced for mopeds.
Type approval process
Mopeds are certified for emission performance according to European regulations. For mopeds
on the road, there are very few safeguards against deterioration of emission performance and
modifications which will affect the emissions. Though, in comparison with current situation, this
issue is improved with the Euro 5 step.
In principle, the sale of mopeds should be supervised by the inspection authority to ensure the
proper functioning of the products sold. However, there is no exact guideline or standard to
check against. Consequently, the inspection is restricted to the documentation with the moped
registration.
Environmental zones
One of the possible policy measures of a local government is the introduction of an environmental
zone. This zone should exclude especially the high emitting mopeds, loud (older) mopeds are
possibly excluded as well. Emission factors of the relevant specific moped classes therefore
became very important. Older and especially older 2-stroke mopeds are known for their high (HC)
emissions. Based on these emission factors, the city of Amsterdam plans to introduce an
environmental zone in 2018 for mopeds with a first registration date before 2011. In this way,
nearly all mopeds with 2-stroke engines will be excluded from the environmental zone of
Amsterdam. More cities in the Netherlands consider environmental zones for mopeds as well.
Electrically powered mopeds seem a good alternative, but are not available on large scale (and
at low costs) as yet.
Future testing methods: In-service-conformity (ISC) and on road measurements
ISC
The measurement results as shown in this paper clearly show the need to control emissions of
in-use vehicles, as some of the emission results are excessively high. These measurements are
performed with Euro 2 mopeds where no durability requirements apply. However, in case
durability and/or CoP requirements are not fully effective at the Euro 5 stage, ISC implementation
can be an effective measure to remedy this problem (Ntziachristos et al., 2017). With these more
stringent limits in Euro 5, the condition of the anti-pollutant devices will become increasingly
important. Ineffective anti-pollution devices can easily cause relatively high exceedances of the
emission limits.

8
On-road measurements
To improve the overall environmental performance assessment in the type-approval procedure,
off-cycle emission testing can be added. A method to assess off-cycle emissions are on road
tests with a Portable Emissions Measurement System (PEMS). By introducing such
requirements, exhaust emissions of more comprehensive driving conditions than the laboratory
test cycle can be assessed. This means a test trip which covers a larger part of the engine load
area, but also different ambient conditions, different road types, etc. In addition, the chance that
vehicles are optimized for a defined emission laboratory test cycle and the risk of cycle beating
significantly decrease when on-road measurement requirements are introduced into the type
approval process.
It is technically feasible to adopt a test procedure for on-road testing by using PEMS
measurements (Ntziachristos et al., 2017). However, the accuracy of the determined mass
emissions is not as high as for passenger cars measured with PEMS.
A regular PEMS, which is commonly used for light- and heavy-duty vehicles, is often too heavy
and too bulky for usage on L-category vehicles, in particular for two-wheeled vehicles. For
example, the typical weight of the main unit only, is often already more than 30 kg. A heavy and
bulky PEMS makes installation more difficult and may influence the driveability and the test
results. Moreover, PEMS is a stand-alone measurement device with a standalone power supply,
which makes the packaging of PEMS even more challenging when a large battery or a generator
is required. A stand-alone power supply is needed because the use of the vehicle’s electric power
output could influence the emission performance. Another complexity is the proper mounting of
an exhaust flow meter. In particular for mopeds, very little space is available around the vehicle
to mount the exhaust flow meter in a proper and safe manner.
For the afore-mentioned reasons, a PEMS with low energy consumption, low weight and limited
dimensions is desirable for mopeds. It should, however, be noted that the following compromises
are often made to make this low weight and low energy consumption possible:
• exhaust flow is not directly measured;
• there are no heated lines;
• the set of analysers differ from a ‘regular’ PEMS. In particular the measurement principle
for HC, NDIR instead of FID, may affect the accuracy.
Since an exhaust flow is required to obtain absolute emissions in grams per kilometre, this PEMS
estimates the flow based on the speed-density method. The inlet air flow can be calculated by
measuring the engine speed, manifold absolute pressure (MAP) and inlet air temperature (IAT).
Then, the exhaust flow can be estimated based on the inlet air flow.
The availability of commercially viable PEMSs suitable for small vehicles, including mopeds, is
rather low in comparison to the light- and heavy-duty vehicles sector (Zardini et al., 2016).
Discussion
With the attention for NOx emissions from diesel vehicles, because of the European NO2 air-
quality standards, other emissions have been put in the background. However, the emissions of
mopeds form a special category, for which automotive measurement standards are less
suitable. The mixture of hydrocarbons and particulates from the tailpipe has been considered by
many as a toxic cocktail, which should deserve proper attention. In particular, with the growing
metropolitan use of mopeds, the exposure of humans to these emissions at street level may
require its own standard. Different parties have looked at different non-regulated aspects of
moped emissions. In Switzerland, the volatile fraction in the particulates is considered most
relevant, and the condensation and accumulation over time has been investigated by, for
example, the PSI (Platt et al., 2014). In the Netherlands health experts considered the exposure
of cyclists to the particulates from the exhaust of mopeds on the bicycle lane in urban
environment (Boogaard & Hoek, 2008). TNO has measured emissions at the side of bicycle
lanes, combined with license plate recognition and speed cameras.
All studies raise the concern for the mixture of particulates and hydrocarbons from the exhaust
gas of mopeds. However, on details the result differs, mainly because of the non-standard
measurement technologies used to address the issue of the exposure of the toxic mixture. They

9
generally assume specific transport process and processes for its health risk. In a simpler
approach, one could say the petrol of a moped is emitted partly unburned, and the composition
of the petrol with about 0.8% benzene, and 25% polyaromatics, is sufficient concern for the
health of cyclists who share the lane with mopeds which can emit up to 10 g/km hydrocarbons.
The cyclists will inhale the gases deeply with the exercise. Whether the pollutants are gases,
droplets, or particulates, is only of secondary concern. In this respect, the total national
emissions may be limited, but still represent a health risk due to the high exposure rates in the
busy urban environment. People use mopeds to avoid road traffic congestion, by using the
bicycle lanes next to the pedestrians and houses.
Due to the more stringent legislation for Euro 4 and 5 mopeds, it is expected that a large part of
such new mopeds will carry more complex and more durable engine technology. Potentially this
should secure a lower level of emissions during the useful life of a vehicle. However, this should
be monitored.
ISC and on-road tests are possible methods to monitor in-use and on-road emission levels of
mopeds. It can be considered to perform ISC testing by application of an on-road test, rather
than the applicable type I test. By combining these two test procedures, the real-world emission
performance of in-use vehicles are thoroughly secured in the most representative way
(Ntziachristos et al., 2017).
Within the ISC requirements for passenger cars, the manufacturer is responsible for the ISC of
the type approved vehicle models. However, when the ISC-programme is performed under full
guidance of the manufacturer, there still is a potential risk that ‘prepared’ vehicles are used.
Instead, representative in-use vehicles deployed in various real-world circumstances, should be
randomly selected. By using representative in-use vehicles, any commonly applied adjustments
at dealerships which can influence the exhaust emissions, are revealed as well. Ideally, the ISC-
verification testing is performed by the TAA independent from the vehicle manufacturer. As a
compromise, TAAs could randomly perform a part of the ISC-verification testing to prevent the
risk of having non-compliant vehicles on a precautionary basis.

10
References
Boogaard, H. and G. Hoek, Blootstelling aan ultrafijnstof tijdens fietsen en autorijden in Nederlandse
steden, 2008, Universiteit Utrecht, P-UB-2008-02.
Eijk, A.R.A., Mensch, P. van, and Cuelenaere, R.F.A., 2016. Brommers in de stedelijke leefomgeving Status
rapport, TNO Report 2016 R10093, Delft, Netherlands.
Ewalds, D., Moritz, G. and Sijstermans, M., Bromfietsen in Nederland, CBS (Statistics Netherlands) report
2013 (Dutch).
Hensema, A., Mensch, P. van, and Vermeulen, R., 2013. Tail-pipe emissions and fuel consumption of
standard and tampered mopeds, TNO Report 2013 R10232, Delft, Netherlands.
Ntziachristos, L., Vonk , W., Papadopoulos, G., van Mensch, P., 2017. Effect study of the environmental
step Euro 5 for L-category vehicles, TNO Report 2017 R10565, The Hague, Netherlands.
Platt, S.M. et al, Two-stroke scooters are a dominant source of air pollution in many cities, Nature
Communications, 13 May 2014.
Zyl, S. van, Van Mensch, P., Ligterink, N., Droge, R. and Kadijk, G., 2015. Update emission model for two-
wheeled mopeds, TNO Report 2014 R11088, Utrecht, Netherlands.
Zardini, A., Clairotte, M., Lanappe, G., Giechaskiel, B. and Martini, G., 2016. Preparatory work for the
Environmental Effect Study on the Euro 5 step of L-category vehicles, European Commission Joint Research
Centre.