Atmospheric Environment 99 (2014) 51e63

Contents lists available at ScienceDirect

Atmospheric Environment
journal homepage: www.elsevier.com/locate/atmosenv

The potential impacts of electric vehicles on air quality in the urban

areas of Barcelona and Madrid (Spain)
A. Soret a, *, M. Guevara a, J.M. Baldasano a, b
a
b

Barcelona Supercomputing Center, Jordi Girona, 29, 08034 Barcelona, Spain

Technical University of Catalonia (UPC), av. Diagonal 647, 08028 Barcelona, Spain

h i g h l i g h t s

Modelling air quality impacts upon EV introduction in Barcelona and Madrid.

EV offers potential air quality improvements, especially related to NO2 and CO.

Lower improvements related to PM due to the high weight of non-exhaust emissions.

A high EV introduction is required (26e40%) to signicantly improve air quality.

Electricity generation emissions due to EV charging imply slight NO2 rises (<3 mg m3).

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 28 January 2014
Received in revised form
16 September 2014
Accepted 18 September 2014
Available online 19 September 2014

This work analyses the potential air quality improvements resulting from three eet electrication
scenarios (~13, 26 and 40%) by replacing conventional vehicles with Electric Battery Vehicles (EBVs),
Plug-in Hybrid Electric Vehicles (PHEVs) and Hybrid Electric Vehicles (HEVs). This study has been performed for the cities of Barcelona and Madrid (Spain), where road transport is the primary emission
source. In these urban areas, several air quality problems are present, mainly related to NO2 and particulate matter. The WRF-ARW/HERMESv2/CMAQ model system has been applied at high spatial
(1  1 km2) and temporal (1 h) resolution. The results show that eet electrication offers a potential for
emission abatement, especially related to NOx and CO. Regarding the more ambitious scenario (~40%
eet electrication), reductions of 11% and 17% of the total NOx emissions are observed in Barcelona and
Madrid respectively. These emissions reductions involve air quality improvements in NO2 maximum
hourly values up to 16%: reductions up to 30 and 35 mg m3 in Barcelona and Madrid, respectively.
Furthermore, an additional scenario has been dened considering electric generation emissions associated with EBVs and PHEVs charging from a combined-cycle power plant. These charging emissions
would produce slight NO2 increases in the downwind areas of <3 mg m3. Thus, eet electrication
would improve urban air quality even when considering emissions associated with charging electric
vehicles. However, two further points should be considered. First, eet electrication cannot be
considered a unique solution, and other management strategies may be dened. This is especially
important with respect to particulate matter emissions, which are not signicantly reduced by eet
electrication (<5%) due to the high weight of non-exhaust emissions. Second, a signicant introduction
of electric vehicles (26e40%) involving all vehicle categories is required to improve urban air quality.
2014 Elsevier Ltd. All rights reserved.

Keywords:
Air quality modelling
Air quality management
Emission inventory
Electric vehicles
Plug-in hybrid electric vehicle

1. Introduction
Air pollution is a major environmental risk to health (WHO,
2011; 2013). A signicant proportion of Europe's population live

* Corresponding author.
E-mail address: albert.soret@bsc.es (A. Soret).
http://dx.doi.org/10.1016/j.atmosenv.2014.09.048
1352-2310/ 2014 Elsevier Ltd. All rights reserved.

in cities, where exceedances of air quality standards occur. In 2011,

42% of trafc stations reported exceedances of the nitrogen dioxide
(NO2) annual limit value (40 mg m3). Regarding particulate matter
(PM) with a diameter up to 10 mm (PM10), the 24-h limit value
(50 mg m3) was exceeded at 43% of trafc sites. In terms of potential to harm human health, the ner PM; up to 2.5 mm (PM2.5) is
the fraction of most concern. The PM2.5 target value threshold
(25 mg m3) was exceeded at 10% of trafc sites (EEA, 2013). In

52

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

Spain, higher NO2 levels occur in major urban areas such as Barcelona and Madrid. The annual limit value was exceeded in both
areas in 2011, while the hourly limit value (200 mg m3) was also
exceeded in Madrid. These areas also recorded air quality problems
related to PM10, while the PM2.5 target value threshold was not
exceeded (MAGRAMA, 2012).
The largest contribution of atmospheric pollutant emissions in
urban areas today is from on-road transport (Colvile et al., 2001;
Belis et al., 2013). In recent years, there have been signicant
efforts to study the effects of strategies designed to reduce onroad trafc emissions and the subsequent impacts of these
emissions on air quality. Currently, the main objectives of these
strategies are either 1) reducing the emission per vehicle by
adopting lower-polluting fuels and technologies (eet renewal by
updating the vehicle emission standards, e.g., Che et al., 2011;
natural gas vehicles, e.g., Gonalves et al., 2009; fuel cell vehicles,
e.g., Stephens-Romero et al., 2009; use of biofuels, (e.g., Liaquat
et al., 2010), or 2) adopting mobility management strategies to
reduce either the maximum speed of circulation (e.g., Baldasano
et al., 2010), or the vehicle kilometres travelled (VKT) (e.g.,
Soret et al., 2011).
Fleet electrication is one of the strategies under consideration
for improving urban air quality. It comprises a wide spectrum of
technology options that range from the early stages of hybrid vehicles to pure electric battery vehicles (EBVs). Hybrid electric vehicles (HEVs) represent the rst step away from a purely
combustion engine vehicle, allowing reduced fuel consumption
compared to conventional gasoline or diesel vehicles (CVs).
Furthermore, plug-in hybrid electric vehicles (PHEVs) can be also
charged from a power grid and can be driven in electric mode over
longer distances and higher speeds than HEVs (Pistoia, 2010).
Finally, EBVs are entirely propelled by stored electricity with no
direct exhaust emissions. Thus EBVs and PHEVs (hereafter referred
to as electric vehicles, EVs) would help to reduce road transport
emissions. Because of the limited driving range in electric mode,
EVs are particularly suitable to improve urban air quality, where
short distances and low speeds are prevalent. Furthermore, higher
potential benets of reducing air emissions are found in highly
populated areas (Ayalon et al., 2013). Depending on the type of
power plant which supplies electric energy for EV, other potential
benets that can be attributable to eet electrication are an increase in energy efciency, and reductions in: energy dependence,
fuel consumption and greenhouse gas (GHG) emissions (ETC/ACC,
2009).
However, EVs entail an additional load on the electricity power
system, resulting in increased emissions from electrical generation,
dependent on power mix. EPRI (2007a,b) forecasted eet electrication in the United States and determined that their electric
demand will cause an increase in coal-red capacity. That study
showed signicant GHG emission reductions. However, modest
effects on nitrogen oxides (NOx) and sulphur dioxide (SO2) were
found, and there was even the possibility of increases in PM in
certain areas. In this way, several studies have explored emission
reductions based on eet electrication, considering various factors
related to the percentage of electric vehicle introduction and the
power source affected by EV demand (e.g., Jansen et al., 2010; Ji
et al., 2012). A clear majority of these studies focus on vehicle
charging at night, coinciding with periods of lower electrical demand; this approach helps to improve the overall utility system
performance and allows for the least expensive electrical production (Parks et al., 2007). Beyond emissions analyses, the number of
studies concerning the air quality impact of eet electrication is
lower, and, in most cases, the spatial resolution used (EPRI, 2007b:
36  36 km2) does not allow for extracting urban air quality conclusions. In other cases, despite the spatial resolution increase (e.g.,

Thompson et al., 2009, 2011; 12  12 and 2  2 km2) only ozone

(O3) variations are analysed.
The main objective of this work is to analyse how eet electrication (introduction of EBVs, PHEVs and HEVs) would reduce
present urban emissions and improve air quality levels (NO2, PM10,
PM2.5, O3, SO2 and CO) in the urban areas of Barcelona and Madrid.
Three eet electrication scenarios are compared with the Base
Case scenario. Furthermore, for the case of Barcelona, an additional
scenario has been dened, considering overnight electric generation emissions associated with EV charging in a natural gas
combined-cycle power plant on the outskirts of Barcelona city. The
state-of-the-art WRF-ARW/HERMESv2/CMAQ model system has
been applied at high spatial (1  1 km2) and temporal (1 h) resolution. Detailed information about on-road trafc and power generation has been collected as input information for the emission
model. This study has been performed for an air pollution episode
(worst-case) that affected the Iberian Peninsula during 2011.
Section 2 presents the applied methodology: modelling system
and study scenarios. Then, Section 3 examines model simulations:
emissions and air quality. Finally, in Section 4, the main conclusions
are discussed.
2. Methodology
2.1. Modelling system
The air quality impacts of eet electrication in the cities of
Barcelona and Madrid were analysed using the Community Multiscale Air Quality (CMAQ) model (Byun and Schere, 2006). The
meteorological elds for CMAQ were generated by the Weather
Research and Forecasting (WRF) meteorological model (Michalakes
et al., 2004; Skamarock and Klemp, 2008) and the High Elective
Resolution Emission Modelling System v2.0 (HERMESv2.0)
(Guevara et al., 2013) provided the emissions for CMAQ. HERMESv2.0 is a high-resolution emission model which uses mainly
bottom-up approaches and Spanish local data.
To obtain adequate boundary and initial conditions, the
modelling system was initially run on two regional scales: the
European domain (12 km  12 km and 1 h) and the Iberian
Peninsula domain (4 km  4 km and 1 h) (Fig. 1a)). A one-way
nesting was performed from one domain to the other to retrieve
the meteorological and chemical conditions to the inner domains.
The nal working domains cover areas of 148  148 km2 and
148  160 km2 (Barcelona and Madrid domains, respectively).
These domains are congured with high horizontal (1 km  1 km)
and temporal (1 h) resolution. 33 s vertical levels are dened, with
12 characterising the planetary boundary layer, for both meteorological and air quality simulations. The top of the model is dened
at 50 hPa to resolve the troposphereestratosphere exchanges
properly. The specications and parameterisations for the WRFARW and CMAQ models are summarised in Table 1. Further information is available in the Supplementary Material.
2.2. HERMESv2.0 and specic emission factors
HERMESv2.0 (Guevara et al., 2013) is a high-resolution emission
model specically developed for Spain; it updates and improves the
original HERMES2004 model (Baldasano et al., 2008). It combines a
comprehensive database with updated methodologies for estimating anthropogenic and biogenic emissions. Given the main
emission sources regarding eet electrication, the specic modules for estimating road transport emissions and combustion processes in energy industries are detailed below.
Combustion emissions from the natural gas combined-cycle
plants have been estimated according to data measured on an

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

53

Fig. 1. Areas of study. 1a) Domains of study: two regional domains (Europe and Iberian Peninsula) and two nal domains (Barcelona and Madrid); 1b) Barcelona nal domain; 1c)
Madrid nal domain; 1d and 1e) detailed images of Barcelona and Madrid cities: Barcelona and Madrid boundaries (red lines), air quality stations (blue points) and natural gas
combined-cycle plants Besos 3e4 (yellow cross).

hourly basis, provided by the Catalan Atmospheric Emission

Network (XEAC, personal communication). To estimate the emissions increases due to the energy demand for EV charging, specic
emissions factors have been calculated from these measured
emissions (Table 2).
Emissions from road transport (exhaust emissions, gasoline
evaporation and road vehicle tyre and brake wear) are estimated
according to the Tier 3 method described in EEA (2009), which is
fully incorporated in version 5.1 of the COPERT 4 software. This tool
offers a comprehensive database with information on emission
factors as a function of the circulation speed for a total of 256
vehicle categories (a comparison of those reported for the most
common vehicles and for a typical speed circulation in urban areas
is performed in Soret et al., 2013). Additionally, HERMESv2.0 also
considers the emissions of particulate matter caused by resuspension from paved road (Pay et al., 2011).
This set of emission methodologies is combined with a digitised
trafc network (Tele Atlas MultiNet, 2011) that contains specic
information by road stretches for trafc intensity (daily average
trafc, [vehicles$day1]) and average circulation speed [km h1].
For the cities of Barcelona and Madrid, this information was obtained by post-processing the combination of trafc intensity from
over 2575 (Barcelona) and 3198 (Madrid) observation stations, and
19 park composition proles based on real circulation data (12 for
Barcelona and 7 for Madrid, respectively).

EEA, (2009) does not describe specic emission factors for EBVs,
PHEVs and HEVs, but specic emission factors have been dened
for this study. First, regarding exhaust and evaporative emissions,
these emissions are regarded as equal to zero for the km travelled in
electric driving mode. For hybrid drive mode, the same emission
factors are assumed for PHEVs and HEVs. For passenger cars (PCs)
driven in hybrid mode, EEA, (2009) denes emission factors for
every pollutant but PM. After a literature review, the authors
decided to include an emission factor of 0.004 g PM km1 (Graham,
2005). As EEA, 2009 considers that the coarse fraction is negligible
in vehicle exhaust emissions, the emission factor of
0.004 g PM km1 is considered PM2.5. For light-duty vehicles (LDVs)
driven in hybrid mode, no emission factors are dened within EEA,
(2009), authors have assumed the same for the Diesel LDV Euro
IV<3.5t. Similarly, for hybrid driven buses the Buses Euro V urban 1518t emission factors are assumed. Finally, regarding non-exhaust
emissions (tyre and brake wear and resuspension), the same
emission factors considered for CVs have been assumed for EBVs,
PHEVs and HEVs.

2.3. Scenarios: eet electrication and energy demand

To study the degree of eet electrication required to reduce the
present air quality problems, three working hypotheses have been
dened: low, medium and high. Table 3 shows the percentage of

54

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

Table 1
Specic parameterizations used in the modelling system for CMAQ and for WRFARW.
Model

Meteorology

Chemistry

Model version
Domains
(Nx,Ny,Nz,
Hor. Res.)
(33 sigma vertical
layers cover
the
troposphere
(up to 50 hPa)
with 12 under
the PBL.
Initial and
boundary
conditions
Parameterizations

WRF-ARW v3
D1: 480, 400, 33,
12  12
D2: 400, 400, 33,
4  4 km
Final Bar.: 148,
148, 33, 1  1 km
Final Mad..: 148,
160, 33, 1  1 km

CMAQ v5.0.1
D1: 480, 400, 15, 12  12
D2: 400, 400, 15, 4  4 km
Final Bar.: 148, 148, 33, 1  1 km
Final Mad..: 148, 160, 33, 1  1 km

Analysis ECMWF
0.5  0.5

D1: LMDz-INCA2
Rest: BC: Parent domain
IC: 24-h Spin-up
Chemical Model: cb5cl-ae5-aq
Boundary l.:YSU
Aerosol Model: AERO5
Microphisics:
Adv scheme: Horiz (Yamartino massWSM3
Cumulus Scheme: conserving)/Vert. (Piecewise Parabolic)
Dif: Eddy diffusivity theory
Kain-Fritsch
Land Surf. Mod.: Aerosol d. v.: aero-depv2
Dry d. r.: Models-3 Cl species
Noah
Long Wave: RRTM
Short wave:
Dudhia

Table 2
Emission factors for a natural gas combined-cycle.

g MWh1

NOx

PM10

CO

VOC

SOx

222.9

1.73

10.3

23.7

18.3

introduction of electric drive mode (EBVs, PHEVs or HEVs) and

hybrid drive mode (PHEVs or HEVs) replacing CVs for each vehicle
category (PCs, LDVs, buses, mopeds and motorcycles).
Fleet electrication has been considered in the cities of Barcelona and Madrid, where the total vehicle-kilometres-travelled
(VKT) are estimated as 13,462,321 and 25,787,145, respectively.
Electric and hybrid drive modes replace a percentage of the current
vehicular eet, maintaining a constant ratio of the current categories of CVs instead of replacing only old vehicles (greatest polluters). It is not in the scope of this work to study eet renewal
(updating Euro standards, updating fuel consumption, etc.) based
on eet electrication.
Fig. 2 shows the percentage of VKT by vehicle category for the
Base Case and for the three hypotheses of eet electrication in
Barcelona and Madrid. The nal percentage of electric and hybrid
drive modes depends on the eet distribution of each city. For
example, an introduction of 30% of electric drive mode in PCs (High
hypothesis) means that the 18.4% of the total VKT in Barcelona will
be accomplished by PCs in electric driving mode. In Madrid, this
percentage is higher (24.4%) because of higher value of the total
VKT of PCs. On the other hand, eet electrication implies a
reduction of the VKT of CVs: e.g., VKT of Diesel PCs (Euro III) are
reduced from 25.1 to 13.8% in Barcelona.
These three hypotheses of eet electrication lead to dene the
scenarios of study:

Base case scenario (BC): current situation (2011), in which no
eet electrication conguration is applied.

Low: a feasible percentage of eet electrication is considered

(~13%). The percentages of VKT in electric drive mode (PCs,
LDVs, buses, mopeds and motorcycles) are 9.7 and 9.1% in Barcelona and Madrid, respectively (Table 4 and Fig. 2). The percentages in hybrid drive mode (PCs, LDVs and buses) are 3.6 and
4.3% in Barcelona and Madrid, respectively.

Medium: in this case, eet electrication is much more ambitious (~26%). VKT in electric drive mode: 19.4 and 18.1%, and in
hybrid drive mode: 7.1 and 8.6% in Barcelona and Madrid,
respectively.

High: signicant electrication is considered (~40%). VKT in
electric drive mode: 29.2 and 27.2%, and in hybrid drive mode:
10.6 and 12.9% in Barcelona and Madrid, respectively.
For the Low, Medium and High scenarios, it is assumed that EV
charging demand is based on overnight electricity generation from
renewables, nuclear or other sources of the Spanish energy system
far from the demand. Thus, it will not imply additional emissions in
the specic fossil-fuel power plants of the selected domains. Nighttime charging allows shorter differences between peak and offpeak demand, improving overall utility system performance
(Thompson et al., 2011), and promoting the generation of energy
using base-load power plants (e.g. nuclear) (ETC/ACC, 2009; Jansen
et al., 2010). Regarding renewable energies, by increasing off-peak
demand, EV can interact synergistically with wind energy
(Hedegaard et al., 2012). The load prole of a night-time charging
regime is well matched to the generating prole of a wind resource,
which also tends to peak at night (T&E, 2009).
To contextualize this assumption within the Spanish electricity
system, it has to be noted that the estimated energy demand for EV
charging (High scenario) represents ~0.4% of total daily energy
demand in Spain (further explained below). Furthermore, the
Spanish electricity demand shows signicant differences between
peak and off-peak demand. Differences between: ~12.000 and
~20.000 MW (REE, 2012). The weight of renewable energy in demand coverage was 33% (16% corresponds to wind power energy),
and 21% of nuclear energy during 2011 (REE, 2012). According to
REE, (2011), by considering smart charging at night-time shorter
differences between peak and off-peak demand would be
observed; improving the overall utility system performance and
promoting the integration of a greater volume of renewable energy,
especially wind energy. Therefore it is feasible to assume a significant contribution of these energy sources (renewables and

Table 3
Percentage of EVs and HEVs introduced for each vehicle category in the cities of
Barcelona and Madrid.
Category

Mode of transport Base case Percentage of EV

and HEV
introduced for each
vehicle category
Low Medium High

Passenger cars (PCs)

Electric
Hybrid
Light-duty vehicles (LDVs) Electric
Hybrid
Heavy-duty vehicles (HDVs) Electric
Hybrid
Buses
Electric
Hybrid
Mopeds
Electric
Motorcycles
Electric

0%
<1%
0%
0%
0%
0%
0%
0%
0%
0%

10%
5%
5%
5%
0%
0%
5%
5%
15%
15%

20%
10%
10%
10%
0%
0%
10%
10%
30%
30%

30%
15%
15%
15%
0%
0%
15%
15%
45%
45%

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

55

Fig. 2. Fraction of VKT for the study scenarios per vehicle category. Vehicle categories that represent <0.1% of the total VKT are not included in the gure.

nuclear) regarding EV charging. A description of the Spanish electricity system is included in the Supplementary Material.
The energy demand for EV charging has been estimated based
on the energy conversion factors for each vehicle category as
described in: the literature, the catalogue of EVs homologated in
Spain and technical descriptions from automobile manufacturers

(Table 5). The same demand is assumed for EBVs and PHEVs for a
selected vehicle category. Furthermore, a conversion efciency ratio from the electrical energy from the grid to power at the wheels
has been considered: 80% (EPRI, 2007b; Berry et al., 2009). According to this, for example, the electric demand for EV charging for
the High scenario is estimated as 909 and 1842 MWh day1 in

56

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

Table 4
Percentage of VKT in electric and hybrid mode considering all the vehicle categories
in Barcelona and Madrid per scenario. And the energy demand estimation due to EV
charging.

% VKT Electric
% VKT Electric
% VKT Hybrid
% VKT Hybrid
Total eet
electrication
MWh day1
MWh day1

City

Low
scenario

Medium
scenario

High and High demand

scenarios

Barcelona
Madrid
Barcelona
Madrid

9.7%
9.1%
3.6%
4.3%
~13%

19.4%
18.1%
7.1%
8.6%
~26%

29.2%
27.2%
10.6%
12.9%
~40%

606
1228

909
1842

Barcelona 303
Madrid
614

Table 5
Energy conversion factors for each vehicle category used to estimate EV charging
demand.
PCs

LDVs

kWh km1 0.2

0.35
Source
Parks et al., 2007 and EPRI, 2007b
ETC/ACC, 2009
and MIET,
2013

Buss

Mopeds Motorcycles

1.26 0.03
BYD, MIET,
2013 2013

0.06
MIET, 2013

Barcelona and Madrid, respectively (Table 4). It represents ~0.4% of

total daily energy demand in Spain.
Furthermore, another scenario has been dened (5 in total) in
order to analyse the higher potential impact of EV charging:

High demand: the same as the High scenario but with energy
demand for EV charging based on available power plants. This
scenario has been dened only for Barcelona because there are
no power plants within the Madrid's domain.
In the High demand scenario, EV demand is based on overnight electricity generation (from 23 to 06 UTC) from a natural gas
combined-cycle plant located in Barcelona (Besos 3e4). This power
plant has been selected, over other power plants located far from
Barcelona, in order to analyse the higher impact due to EV charging.
Fig. 3 shows the hourly electricity generated during the study
period (3e5 October, 2011) and the estimated increase due to the
EV charging.
2.4. Selection of the studied period (3e5 october, 2011)
The air quality impacts of eet electrication have been analysed for a critical episode of air pollution affecting the entire Iberian Peninsula: 3e5 October, 2011 (worst-case). This period has
been selected based on air quality monitoring data but also
considering weekdays in order to focus the study on the usual
trafc circulation patterns.
From the synoptic point of view, the episode of 3e5 October,
2011 is characterised by a high pressure system situated over the
Iberian Peninsula. This induced a western recirculation over the
Iberian Peninsula (see supplementary material). These conditions
dominate 45% of the annual transport pattern over the northeastern Iberian Peninsula and 36% in the central Iberian Peninsula
(Jorba et al., 2004). These situations lead to weak synoptic forcing.
Thus, mesoscale phenomena, induced by the topography of each
region, dominate circulation patterns.
Barcelona is situated in the north-eastern Iberian Peninsula, in a
coastal area characterised by a very complex terrain. The presence
of sea breezes and the development of a thermal boundary layer
induce the layering and accumulation of pollutants (Mill

an et al.,

1997; Perez et al., 2004). During the period of study, a welldeveloped land-sea breeze regime dominated the local atmospheric dynamics (see supplementary material). Madrid is located
in the centre of the Iberian Peninsula, where convective thermal

n et al., 1997).
phenomena control the pollutant dynamics (Milla
These atmospheric processes affected the pollutant dynamics
involving their accumulation. The primary air quality problems
during the study period were related to NO2 and PM10. The NO2
hourly limit value (200 mg m3) was exceeded in Barcelona and
Madrid. PM10 daily levels exceed the 24-h limit value in Barcelona
(50 mg m3), while in Madrid those levels were close to the limit
value. Regarding secondary pollutants, the solar radiation and
temperature are lower in October than in summer, thus reducing
photochemical reactions. O3 hourly levels were under 120 mg m3
(the population information threshold is 180 mg m3).
3. Results
In this section, emission and air quality model results of the
1 km  1 km simulations are presented. The WRF-ARW/
HERMESv2/CMAQ modelling system has been evaluated for the
available air quality stations of Barcelona (7) and Madrid (19)
(Fig. 1d) and e)). The model evaluation is shown in the
Supplementary Material.
3.1. Emissions
3.1.1. Base case scenario; current situation
In the cities of Barcelona and Madrid, road transport is the main
emission source for NOx, PM10 PM2.5 and CO. Table 6 shows the
daily average emissions for the period of study. In Madrid, trafc
emissions represent 69% of the total NOx emissions; in Barcelona,
this percentage is lower: 40% (Fig. 4a and b). In Barcelona, port and
energy industries emissions have a signicant weight, representing
the 29 and the 13% of the total NOx emissions, respectively.
Regarding PM10 emissions, the contribution of trafc emissions
is much more signicant: 80 and 90% for Barcelona and Madrid,
respectively (77 and 89% concerning PM2.5, respectively). PM10
exhaust emissions represent ~40% of the total road trafc emissions
(~30% regarding PM2.5). The rest is due to non-exhaust emissions:
resuspension and brake, tyre and road abrasion.
Fig. 4c, d, e and f compare the VKT per vehicle categories and the
corresponding road transport emissions. While the VKT travelled
by passenger cars represent 61 and 81% of the total VKT, the relative
weights of these emissions are lower: 38 and 43% of the total NOx
emissions in Barcelona and Madrid, respectively. Other means of
transport are greater polluters. For example, in Madrid, the VKT of
buses and heavy-duty vehicles (HDVs) represent 9% of the total
VKT, but their NOx emissions are 49% of the total NOx trafc
emissions. This also affects the rest of the pollutants. In Barcelona,
with a large number of two-wheeled vehicles, CO emission from
motorcycles represents 64% of total transport CO emissions, while
its VKT in only 12%.
3.1.2. Fleet electrication scenarios: low, medium, high scenarios
The emission reductions are described below for the road
transport sector and for the total emissions. The analysis is focused
on the differences between the Base case and High (~40% eet
electrication) scenarios. However, the differences of the Low
(~13%) and Medium (~26%) scenarios are proportional.
Fleet electrication implies emission reductions in all the pollutants analysed. These reductions depend on three main factors:

Percentage of electrication considered: Low, Medium and
High.

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

57

Fig. 3. Hourly electricity generation of the natural gas combined-cycle plants Besos 3e4 for the period of study (3e5 October, 2011), with increases due to overnight electric
generation associated with EV charging.

Relative weight of the exhaust road transport emissions to the

total emissions.

Vehicular eet composition.
NOx trafc reductions are similar in both cities: 27 and 25% in
Barcelona and Madrid, respectively. The differences are related to
the vehicular eet composition of each city: LDVs are more common in Barcelona, while Madrid has a higher percentage of HDVs (a
further description of how eet electrication reduces road transport emissions by vehicle category is shown in the Supplementary
Material). However, the total NOx reduction is lower in Barcelona
(11%) than in Madrid (17%), due to the higher weights of other
sources (Table 6, and Fig. 4g and h).
The reductions in PM10 trafc emissions are comparable for
Barcelona (5%) and Madrid (3%). These percentages of reductions
are lower than for other pollutants because eet electrication does
not reduce non-exhaust emissions. The differences between Barcelona and Madrid are due to the eet composition; the higher
percentage of HDVs in Madrid diminishes the impact in PM10
emission reductions. Because road transport is the main emission
source of PM10, the achieved reductions in transport are similar to
the total emissions reductions: 4 and 3%, respectively. Regarding
PM2.5, somewhat higher trafc emission reductions are observed (7
and 5% for Barcelona and Madrid, respectively) because exhaust
road transport emits the ne fraction of PM, which implies higher
total PM2.5 emissions reductions (5% in both cities).
CO trafc emission reductions are estimated as 41 and 38% for
Barcelona and Madrid, respectively. Most of the reduction of CO
emissions in Barcelona is due to the introduction of electric motorcycles. Similar to CO, NMVOC reductions of road transport in
Barcelona are higher than in Madrid: 23% and 12%, respectively, due
to the introduction of electric mopeds. Concerning SOx, trafc
emission reductions are high: 26 and 27% for Barcelona and Madrid,
respectively. However, its impact on the total emission reduction is
very low (1 and 6%, respectively) because road transport is not the
main emission source of SOx.

3.1.3. High demand scenario

Comparison of the High and High demand scenarios shows
that the increased electrical generation will not entail signicant
increases in any of the pollutants analysed (<1%) (Table 6 and
Fig. 4g). The higher increase is observed with respect NOx emissions
(increase of 0.20 Mg day1). For the rest of the pollutants lower
increases are observed (0.02 Mg day1 for SOx and NMVOC,
0.01 Mg day1 for CO, and 0.002 Mg day1 for PM10 and PM2.5).
Thus, comparing Base Case and High demand scenarios, eet
electrication implies a net reduction of the total emissions.
3.2. Air quality
3.2.1. Base case scenario
Model results show that the NO2 hourly limit value (200 mg m3)
was exceeded in downtown Barcelona and Madrid (Fig. 5a and b).
The maximum 24-h level of NO2 also exceeded the annual limit
value (40 mg m3) in both urban areas.
Fig. 5c and d show the spatial distribution of the PM10 maximum
hourly levels in Barcelona and Madrid, respectively. In Barcelona,
the maximum hourly levels were observed in downtown Barcelona, in the industrial areas and in the accessing roads to Barcelona.
In these areas, the 24-h average exceeded the target level
(50 mg m3). In Madrid, the 24-h target reached the 45 mg m3, close
to the target level. In both areas (Barcelona and Madrid), these 24-h
levels exceed the annual limit target (40 mg m3). Concerning ner
particles, the 24-h average of PM2.5 (particulate matter with a
diameter up to 2.5 mm) reached 40 mg m3 in both cities, which is
higher than the annual target value threshold is 25 mg m3.
In both Barcelona and Madrid, O3 levels are lower downtown
(maximum hourly concentration <85 mg m3) than in the downwind areas (>85 mg m3). This behaviour is due to the higher
concentration of fresh NO that acts as an O3 sink, as well as to the
depletion of radicals via HNO3 formation by NO2 consumption
(Atkinson, 2000). In the areas of study, O3 levels did not exceed
either the hourly threshold of 180 mg m3 or the 8-h maximum
target of 120 mg m3. Regarding SO2, 24-h concentrations are far
from both the 24-h limit value (125 mg m3) and the hourly limit

58

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

Table 6
NOx, PM10, PM2.5, CO, SOx and NMVOC estimated emissions by sector for the EB scenario for the cities of Barcelona and Madrid, and the total emissions for the other scenarios
and the corresponding increase compared to the Base case scenario.
Mg day1

Barcelona city
NOx

Combustion in energy industries (SNAP01)

4.13
Non-industrial combustion (SNAP02)
3.20
Combustion & Process in Industries (SNAP03/04)
1.76
Distribution of fossil fuels (SNAP05)
0.00
Solvents (SNAP06)
0.00
Road transport (SNAP07)
12.57
Other mobile sources (SNAP08)
9.06
Waste treatment (SNAP09)
0.69
Agriculture (SNAP10)
0.00
Biogenic emissions (SNAP11)
0.01
Total Base Case scenario
31.42
Total Low scenario
30.30
Total Medium scenario
29.18
Total High scenario
28.06
Total High demand scenario
28.26
D Low e EB
4%
D Medium e EB
7%
D High e EB
11%
D High demand e EB
10%

Madrid city

PM10

PM2.5

CO

0.17
0.07
0.03
0.00
0.00
2.44
0.34
0.01
0.00
0.00
3.06
3.02
2.98
2.95
2.95
1%
3%
4%
4%

0.17
0.06
0.03
0.00
0.00
1.91
0.32
0.01
0.00
0.00
2.50
2.46
2.41
2.37
2.37
2%
3%
5%
5%

1.80
0.22
1.71
0.23
0.95
0.09
0.00
0.00
0.00
0.00
28.07
0.19
0.97
3.75
0.17
0.04
0.00
0.00
0.12
0.00
33.79
4.52
29.95
4.50
26.10
4.49
22.26
4.47
22.26
4.49
11%
0%
23%
1%
34%
1%
34%
1%

value (350 mg m3) in both cities. The higher hourly levels are
located in Barcelona, around the port (<65 mg m3) due to the port
emissions. Concerning CO, the maximum 8-h levels in Madrid and
Barcelona were <3 mg m3, far from the 8-h limit value of
10 mg m3. Figures for O3, SO2 and CO are not shown.
3.2.2. Fleet electrication scenarios: low, medium and high
scenarios
Fig. 5eeh shows the differences in the maximum hourly levels
for NO2 and PM10 between the Base Case and High (~40% eet
electrication) scenarios. Higher pollutant concentration abatements were observed in downtown Barcelona and Madrid, where
eet electrication is considered, and to a lesser extent in the
corresponding downwind areas. The higher NO2 reductions were
<30 mg m3 in downtown Barcelona and <35 mg m3 in downtown
Madrid. PM10 reductions are smaller than NO2 reductions:
<8 mg m3 in downtown Barcelona and <6 mg m3 in downtown
Madrid.
In contrast, the O3 hourly maximum concentrations increase
with eet electrication: <4 mg m3. This behaviour is characteristic of VOC-sensitive areas, usually in conditions with low VOC to
NOx ratios (Sillman and He, 2002). It should be noted that these O3
increases occur where the concentrations in the Base Case scenario
are lower (inner area of both cities). Regarding SO2, hourly
maximum reductions are not signicant; <2 mg m3 in both
downtown areas. Concerning CO, hourly maximum reductions are
higher in Barcelona (0.5 mg m3) than in Madrid (<0.3 mg m3).
Figures for O3, SO2 and CO are not shown.
Furthermore, the hourly air quality impacts of eet electrication has been analysed for the locations with air quality station in
Barcelona (7) and Madrid (19). Fig. 6 shows the hourly concentrations for NO2, PM10, and PM2.5 for the Base case, Low, Medium and
High scenarios. Two urban-trafc stations for each city were
selected: Eixample (41.386N; 2.155E) and Poblenou (41.405N;
2.205E) in Barcelona, and Castellana (40.439N; 3.690W) and
Escuelas Aguirre (40.421N; 3.682W) in Madrid. Measured air
quality levels are also shown to be compared with the Base case
scenario results, to show model performance. Further model evaluation is shown in the Supplementary Material.
In both Madrid and Barcelona, higher NO2 reductions are
observed for the higher air quality levels. For the Eixample station,

SOx

NMVOC NOx
0.48
0.52
0.45
1.03
12.16
10.41
0.29
0.01
0.00
0.70
26.05
25.25
24.46
23.66
23.68
3%
6%
9%
9%

0.00
4.66
0.91
0.00
0.00
28.43
7.26
0.00
0.00
0.02
41.27
38.90
36.52
34.13
e
6%
12%
17%
e

PM10
0.00
0.20
0.09
0.00
0.00
5.32
0.29
0.00
0.00
0.00
5.89
5.83
5.77
5.71
e
1%
2%
3%
e

PM2.5
0.00
0.17
0.06
0.00
0.00
4.15
0.28
0.00
0.00
0.00
4.66
4.59
4.52
4.45
e
2%
3%
5%
e

CO

SOx
0.00
2.64
8.24
0.00
0.00
25.60
8.32
0.00
0.01
0.27
45.08
41.94
38.66
35.38

e
7%
14%
22%
e

0.00
0.88
0.11
0.00
0.00
0.41
1.57
0.00
0.00
0.00
1.93
1.90
1.86
1.82
e
2%
4%
6%
e

NMVOC
0.00
0.80
0.18
1.00
15.87
11.36
3.85
0.00
0.01
1.67
32.17
31.73
31.29
30.84
e
1%
3%
4%
e

the higher level observed for the Base case scenario is reduced by
26 mg m3 (from 194 to 168 mg m3), which represents a 13%
reduction. In Barcelona, the reductions of the maximum levels are
between 8 and 16%. The reductions of the 24-h values are lower:
between 7 and 13%. In Madrid, higher reductions are observed for
the 24-h average; 11e15% and similar for the hourly maximums:
8e16%. In Madrid, higher air quality improvements regarding the
NO2 24-h average are obtained because the contribution of road
transport to the total emission is higher than in Barcelona. In Barcelona, the plume from the port and from energy industries plays
an important role in the air quality levels depending on local wind
patterns.
Smaller reductions were observed for PM10 values. In Barcelona,
the higher hourly values were reduced by 3e5%; in Madrid, these
reductions were 2e4%. Regarding the ne fraction of PM, PM2.5
levels were reduced more than PM10 (4e7% and 3e5% for the
higher values in Barcelona and Madrid, respectively).
3.2.3. High demand scenario
Fig. 7 shows the air quality impacts of EV charging demand
based on overnight electricity generation from a gas combinedcycle plant in Barcelona (909 MWh day1). At night, land breeze
directs the combined-cycle plume seaward. These results show
slight NO2 increases in the downwind areas <3 mg m3 (Fig. 7a). For
the other pollutants, PM10, PM2.5, O3, SO2 and CO, the increases are
not signicant (<1 mg m3). Thus, signicant air quality benets are
observed with eet electrication, even considering the electric
generation emissions associated with EV charging. This is based on
a reduction in the total emissions and an increase of the dispersive
features of these emissions.
4. Conclusions
This work analyses the potential emission and air quality impacts of eet electrication in Barcelona and Madrid. The conclusions shown in this study are derived from the study of an air
pollution episode in 2011 (worst-case). Although it cannot be
considered a large period, it is an episode with signicant air
pollution. The study was performed by applying the WRF-ARW/
HERMESv2/CMAQ model system at high spatial (1  1 km2) and
temporal (1 h) resolution. This level of detail allowed for the

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

59

Fig. 4. Daily average emissions for the period of study for the cities of Barcelona (left panels) and Madrid (right panels). (Upper panels) 4a and 4c; NOx emissions per sector, 4b and
4d; total emissions per scenario. (Lower panels) Percentage of VKT by vehicle category (4e and 4g) and the corresponding road transport emissions (4f and 4h).

60
A. Soret et al. / Atmospheric Environment 99 (2014) 51e63
Fig. 5. Air quality levels in Barcelona and Madrid for the period of study (3e5 October, 2011). Upper panels: Barcelona; Lower panels: Madrid. From left to right: maximum hourly levels of NO2 (5a and 5b), maximum hourly levels of
PM10 (5c and 5d), NO2 maximum hourly differences between Base Case and High scenario (5e and 5f) and PM10 maximum hourly differences (5g and 5h). Roads (grey lines) and Barcelona and Madrid area boundaries (black lines) are
also included.

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

Fig. 6. Hourly air quality levels in Barcelona and Madrid for the period of study (3e5 October, 2011) for the Base Case (red dotted line), Low (yellow dotted line), Medium (green dotted line), High (blue dotted line) scenarios, observations
(black dots), and limit values and thresholds (red lines). From left to right: NO2, PM10 and PM2.5 and from top to bottom: Poblenou (Barcelona), Eixample (Barcelona), Castellana (Madrid) and Escuelas Aguirre (Madrid) stations.

61

62

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

Fig. 7. Air quality differences between High demand and High scenarios of NO2 (7a) and PM10 (7b) in Barcelona. Roads (grey lines), natural gas combined-cycle location (red dot)
and Barcelona and Madrid boundaries (black lines) are also included.

analysis of urban air quality. Three eet electrication scenarios

(~13, 26 and 40%), considering the introduction of EBVs, PHEVs and
HEVs, have been dened for contrast with a Base case scenario in
which no eet electrication conguration is applied. Furthermore,
and only for the case of Barcelona, an additional scenario has been
dened, considering overnight electric generation emissions associated with electrical vehicles charging from a natural gas
combined-cycle plant.
In terms of emissions, these results show that eet electrication offers a potential for NOx and CO abatement. Regarding the
more ambitious scenario (~40%), reductions of 11% and 17% of the
total NOx emissions are observed in the cities of Barcelona and
Madrid, respectively. These emission reductions involve air quality
improvements in NO2 of 8e16% for the maximum hourly values:
reductions up to 30 and 35 mg m3 in Barcelona and Madrid,
respectively. Moreover, results obtained considering overnight
electric generation emissions do not show signicant increases for
any pollutant in the city. Only slight NO2 increases in the downwind
areas <3 mg m3 are observed.
In contrast, eet electrication has a limited impact on PM10
reductions in terms of both emissions (3e4%) and air quality
(2e5%). This is due to the high contribution of non-exhaust emissions (resuspension and brake, tyre and road abrasion), which
cannot be reduced by eet electrication. Regarding PM2.5, somewhat higher emissions reductions are observed because exhaust
road transport emits the ne fraction of PM, which implied slightly
higher air quality improvements (3e7%).
Moreover, O3 hourly maximum concentrations increase (<4% for
the maximum hourly and 8-h levels) in both downtown areas, due
to the reduction of NO, which acts as an O3 sink.
In general, eet electrication presents potential air quality
benets. However, certain factors must be considered. Although
eet electrication leads to signicant improvements, the reductions achieved are insufcient to ensure proper air quality
levels. This suggests that electrication cannot be considered the
sole solution, especially regarding particulate matter. Additionally,
the percentage of eet electrication required (26e40%) to acquire
a signicant level of improvement suggests that all the stakeholders must work together in order to assess such paradigm shift.

This also suggests that eet electrication will have a limited

impact on air quality for at least the next decade. Moreover, eet
electrication and other management measures should involve all
vehicle categories (two-wheelers, heavy-duty vehicles, buses and
light-duty vehicles) not only passenger cars. The contribution to the
total road transport emissions of heavy-duty vehicles and buses is
signicant, for example, it represents the 49% of the total NOx
transport emissions in Madrid. Thus, road complementary transport management strategies must be dened so that their emissions can be further reduced. Finally, although road transport is the
main emission source in urban areas, further emission reductions
should be considered for the other sources. For example, in the case
of Barcelona, the weights of the port and the energy industry play
important roles in the air quality levels.
In conclusion, this work has proved that eet electrication
would improve urban air quality, even when assuming that the EV
demand is based on a combined-cycle plant located at the periphery of the city. However, the air quality impacts of electricity
generation to power EV are not limited to urban areas. If, for
instance, coal-red power plants allocated in isolated areas are
used to generate electricity to power EV, SO2 emissions will increase in that location, then more particulate sulphate would be
formed and transported, with the possibility of affecting urban
areas. In this sense, secondary pollutants dynamics and other implications of eet electrication will be assessed in future works.
Finally, we conclude that the WRF-ARW/HERMESv2/CMAQ
modelling system is a suitable tool for the management and
assessment of urban air quality. The air quality results obtained for
the Base case scenario have been extensively evaluated with
observational data. Furthermore, air quality improvements for
primary pollutants are consistent with dened emission
reductions.

Acknowledgement
The Authors gratefully acknowledge K. Serradell for the meteorological input data, F. Martinez for the implementation of HERMESv2.0 and J. Caus from SEAT for his comments and guidance
regarding electried vehicles. Authors also want to thank the

A. Soret et al. / Atmospheric Environment 99 (2014) 51e63

anonymous reviewers whose comments helped to improve this

paper substantially. This work has been written with the support of
the grant SEV-2011-00067 of Severo Ochoa Program, awarded by
the Spanish Government. Simulations were carried out on the
MareNostrum supercomputer of the Barcelona Supercomputing
CentereCentro Nacional de Supercomputacion.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.atmosenv.2014.09.048.
References
Atkinson, R., 2000. Atmospheric chemistry of VOCs and NOx. Atmos. Environ. 34,
2063e2101. http://dx.doi.org/10.1016/S1352-2310(99)00460-4.
Ayalon, O., Flicstein, B., Shtibelman, A., 2013. Benets of reducing air emissions:
replacing conventional with electric passenger vehicles. J. Environ. Prot. 4,
1035e1043. http://dx.doi.org/10.4236/jep.2013.410119.

pez, E., Gasso

, S., Jime

nez-Guerrero, P., 2008.
Baldasano, J.M., Gereca, P., Lo
Development of a high resolution (1 km  1 km, 1 h) emission model for Spain:
the High-Elective Resolution Modelling Emission System (HERMES). Atmos.
Environ. 42, 7215e7233. http://dx.doi.org/10.1016/j.atmosenv.2008.07.026.

nez-Guerrero, P., 2010. Air pollution
Baldasano, J.M., Gonalves, M., Soret, A., Jime
impacts of speed limitation measures in large cities: the need for improving
trafc data in a metropolitan area. Atmos. Environ. 44, 2997e3006. http://dx.
doi.org/10.1016/j.atmosenv.2010.05.013.
Belis, C.A., Karagulian, F., Larsen, B.R., Hopke, P.K., 2013. Critical review and metaanalysis of ambient particulate matter source apportionment using receptor
models in Europe. Atmos. Environ. 69, 94e108. http://dx.doi.org/10.1016/j.
atmosenv.2012.11.009.
Berry, I., Khusid, M., Kasseris, M., Mak, A., 2009. What's the Deal with Hybrid and
Electric Cars? MIT Electric Vehicle Team. http://web.mit.edu/evt/EVT2009_
IAPClass_Day1.pdf. January 2014.
BYD, 2013. Electric Bus, Economy. Build your dreams. http://www.byd.com/na/auto/
ElectricBus.html#p2. January 2014.
Byun, D.W., Schere, K.L., 2006. Review of the governing equations, computational
algorithms and other components of the Models-3 Community Multiscale Air
Quality (CMAQ) modeling system. Appl. Mech. Rev. 59, 51e77. http://dx.doi.
org10.1115/1.2128636.
Che, W., Zheng, J., Wang, S., Zhong, L., Lau, A., 2011. Assessment of motor vehicle
emission control policies using Model-3/CMAQ model for the Pearl River Delta
region, China. Atmos. Environ. 45, 1740e1751. http://dx.doi.org/10.1016/j.
atmosenv.2010.12.050.
Colvile, R.N., Hutchinson, E.J., Mindell, J.S., Warren, R.F., 2001. The transport sector
as a source of air pollution. Atmos. Environ. 35, 1537e1565. http://dx.doi.org/10.
1016/S1352-2310(00)00551-3.
EEA, 2009. EMEP-CORINAIR Emission Inventory Guidebook e 2009. European
Environmental Agency Technical Report N.9. http://www.eea.europa.eu/
publications/emep-eea-emission-inventory-guidebook-2009. January 2014.
EEA, 2013. Air Quality in Europe d 2013 Report. European Envirionmental Agency,
ISBN 978-92-9213-406-8. No 9/2013. http://dx.doi.org/10.2800/92843.
EPRI, 2007a. Environmental Assessment of Plug-in Hybrid Electric Vehicles. In:
Nationwide Greenhouse Gas Emissions, vol. 1. EPRI, Palo Alto, CA, 2007.
1015325.
EPRI, 2007b. Environmental Assessment of Plug-in Hybrid Electric Vehicles. In:
United States Air Quality Analysis Based on aeo-2006 Assumptions for 2030,
vol. 2. EPRI, Palo Alto, CA, 2007. 1015325.
ETC/ACC, July 2009. Environmental Impacts and Impact on the Electricity Market of
a Large Scale Introduction of Electric Cars in Europe - Critical Review of Literature. European Topic Centre for Air Pollution and Climate Change Mitigation,
Berlin. Technical Paper 2009/4.

nez-Guerrero, P., Baldasano, J.M., 2009. High resolution
Gonalves, M., Jime
modeling of the effects of alternative fuels use on urban air quality: introduction of natural gas vehicles in Barcelona and Madrid Greater Areas (Spain). Sci.
Total Environ. 407, 776e790. http://dx.doi.org/10.1016/j.scitotenv.2008.10.017.

valo, G., Gasso

, S., Baldasano, J.M., 2013. An improved
Guevara, M., Martnez, F., Are
system for modelling Spanish emissions: HERMESv2.0. Atmos. Environ. 81,
209e221. http://dx.doi.org/10.1016/j.atmosenv.2013.08.053.
Graham, L., 2005. Chemical characterization of emissions from advanced technology light-duty vehicles. Atmos. Environ. 39, 2385e2398. http://dx.doi.org/10.
1016/j.atmosenv.2004.10.049.
Hedegaard, K., Ravn, H., Juul, N., Meibom, P., 2012. Effects of electric vehicles on
power systems in Northern Europe. Energy 48 (1), 356e368.
Jansen, K.H., Brown, T.M., Samuelsen, G.S., 2010. Emisssions impacts of plug-in
hybrid electric vehicle deployment on the US Western grid. J. Power Sources
195, 5409e5416. http://dx.doi.org/10.1016/j.jpowsour.2010.03.013.

63

Ji, S., Cherry, C.R., Bechle, M.J., Wu, Y., Marshall, J.D., 2012. Electric vehicles in China:
emissions and health impacts. Environ. Sci. Technol. 46, 2018e2024. http://
pubs.acs.org/doi/pdf/10.1021/es202347q.

rez, C., Rocadenbosch, F., Baldasano, J.M., 2004. Cluster analysis of 4-day
Jorba, O., Pe
back trajectories arriving in the Barcelona area (Spain) from 1997 to 2002.
J. Appl. Meteorol. 43, 887e901. http://dx.doi.org/10.1175/1520-0450(2004)
043<0887:CAODBT>2.0.CO;2.
Liaquat, A.M., Kalam, M.A., Masjuki, H.H., Jayed, M.H., 2010. Potential emissions
reduction in road transport sector using biofuel in developing countries. Atmos.
Environ. 44, 3869e3877. http://dx.doi.org/10.1016/j.atmosenv.2010.07.003.
MAGRAMA, November 2012. Air Quality Assessment of Spain for 2011. Environmental ministry of Spain, p. 43.
MIET, March 2013. Electric Vehicle Catalogue Homologated in Spain. Industry, energy and turism ministry. http://www.movele.es/index.php/mod.coches/mem.
listado/relmenu.4. October 2013.
Michalakes, J., Dudhia, J., Gill, D., Henderson, T., Klemp, J., Skamarock, W., et al.,
2004. The weather research and forecast model: software architecture and
performance. In: Mozdzynski, E.G. (Ed.), To Appear in Proceeding of the Eleventh ECMWF Workshop on the Use of High Performance Computing in Meteorology, 25e29 October 2004, Reading, U.K.

n, M., Salvador, R., Mantilla, E., Kallos, G., 1997. Photo-oxidant dynamics in the
Milla
Mediterranean basin in summer: results from European research projects.
J. Geophys Res. 102, 8811e8823. http://dx.doi.org/10.1029/96JD03610.
Parks, K., Denholm, P., Markel, T., 2007. Costs and Emissions Associated with Plug-in
Hybrid Electric Vehicle Charging in the Xcel Energy Colorado Service Territory.
Tech. Rep. NREL/TP-640e41410. National Renewable Energy Laboratory.

nez-Guerrero, P., Baldasano, J.M., 2011. Implementation of resusPay, M.T., Jime
pension from paved roads for the improvement of CALIOPE air quality system in
Spain. Atmos. Environ. 45, 802e807. http://dx.doi.org/10.1016/j.atmosenv.2010.
10.032.

rez, C., Sicard, M., Jorba, O., Comero

n, A., Baldasano, J.M., 2004. Summertime rePe
circulations of air pollutants over the north-eastern Iberian coast observed from
systematic EARLINET lidar measurements in Barcelona. Atmos. Environ. 38,
3983e4000. http://dx.doi.org/10.1016/j.atmosenv.2004.04.010.
Pistoia, G., 2010. Electric and Hybrid Vehicles. Elsevier, Amsterdam, ISBN
9780444535658, p. 670.
REE, 2011. Electric vehicle. Spanish electric operator vision. In: Genera 2011 International Conference on Energy and Envirionment, Madrid, Spain, 11e12 May
2011. http://www.ree.es/sites/default/les/downloadable/genera2011.pdf. July
2014.

ctrica Espan
~ ola, p. 147.
REE, 2012. The Spanish Electricity System 2011. Red Ele
http://www.ree.es/en/publications/2013/10/spanish-electricity-system-report2011. November 2013.
Sillman, S., He, D.Y., 2002. Some theoretical results concerning O3-NOx-VOC
chemistry and NOx-VOC indicators. J. Geophys. Res. Atmos. 107, 2156e2202.
http://dx.doi.org/10.1029/2001JD001123.
Skamarock, W.C., Klemp, J.B., 2008. A time-split nonhydrostatic atmospheric model
for weather research and forecasting applications. J. Comput Phys. 227,
3465e3485. http://dx.doi.org/10.1016/j.jcp.2007.01.037.
Soret, A., Jimenez-Guerrero, P., Baldasano, J.M., 2011. Comprehensive air quality
planning for the Barcelona Metropolitan Area through trafc management.
Atmos. Pollut. Res. 2, 255e266. http://dx.doi.org/10.5094/APR.2011.032.

nez-Guerrero, P., Andre

s, D., Ca

rdenas, F., Rueda, S., Baldasano, J.M.,
Soret, A., Jime
2013. Estimation of future emission scenarios for analysing the impact of trafc
mobility on a large Mediterranean conurbation in the Barcelona Metropolitan
Area (Spain). Atmos. Pollut. Res. 4, 22e32. http://dx.doi.org/10.5094/APR.2013.
003.
Stephens-Romero, S., Carreras-Sospedra, M., Brouwer, J., Dabdub, D., Samuelsen, S.,
2009. Determining air quality and greenhouse gas impacts of hydrogen infrastructure and fuel cell vehicles. Environ. Sci. Technol. 43, 9022e9029. http://dx.
doi.org/10.1021/es901515y.
TeleAtlasMultinet, 2011. European and Spanish Digital Cartography. http://www.
tomtom.com/en_gb/licensing/products/maps/multinet/. January 2014.
T&E, November 2009. How to Avoid an Electric Shock-Electric Cars, from Hype to
Reality. European Federation for Transport and Environment AISBL, Brussels.
Thompson, T.M., Webber, M., Allen, D.T., 2009. Air quality impacts of using overnight electricity generation to charge plug-in hybrid electric vehicles for daytime use. Environ. Res. Lett. 4, 014002. http://dx.doi.org/10.1088/1748-9326/4/
1/014002.
Thompson, T.M., King, C.W., Allen, D.T., Webber, M.E., 2011. Air quality impacts of
plug-in hybrid electric vehicles in Texas: evaluating three battery charging
scenarios. Environ. Res. Lett. 6, 024004. http://dx.doi.org/10.1088/1748-9326/6/
2/024004.
WHO, 2011. Air Quality and Health d Fact Sheet No 313. World Health Organization. Updated September 2011. http://www.who.int/mediacentre/factsheets/
fs313/en/. January 2014.
WHO, 2013. Review of Evidence on Health Aspects of Air Pollution - REVIHAAP
Project. First Results. WHO Regional Ofce for Europe, Copenhagen, Denmark,
p. 33.