To what extent the traffic restriction policies applied in Barcelona
city can improve its air quality?
Daniel Rodriguez-Reya,∗, Marc Guevaraa, Mª Paz Linaresb, Josep Casanovasa,b, Jan M.
Armengola, Jaime Benavidesa,d, Albert Soreta, Oriol Jorbaa, Carles Tenaa, Carlos Pérez
Garćıa-Pandoa,c
Barcelona Supercomputing Center, Barcelona, 08034, Spain
aBarcelona Supercomputing Center, Barcelona, 08034, Spain
bUniversitat Politècnica de Catalunya-Barcelona Tech UPC – Carrer Jordi Girona 1-3 08034 Barcelona
cICREA, Catalan Institution for Research and Advanced Studies, Barcelona, 08010, Spain
dNow at Department of Environmental Health Sciences, Mailman School of Public Health, Columbia
University, New York NY 10032, USA.
Abstract
Barcelona city (Spain) is applying a series of traffic restriction measures that aim at renewing
and reducing the amount of circulating vehicles to improve air quality. The measures include
changes in the built environment to reduce private vehicle space in specific areas through the
so-called ”superblocks” and tactical urban planning actions, along with the implementation
of a city-wide Low Emission Zone (LEZ) that restricts the entry of the most polluting
vehicles to the city. Our study quantifies the impact of these measures in the greater area
of Barcelona combining a coupled macroscopic traffic and pollutant emission model with a
multi-scale air quality model. Our modelling system allows estimating the effect of different
traffic restrictions upon traffic and the associated emissions and air quality levels at a very
high resolution (20 m). The measures were evaluated both individually and collectively
to assess both their relative and overall impact upon emissions and air quality. We show
that in the absence of traffic demand reductions, the application of isolated measures that
reduce private vehicle space, either through superblocks or tactical urban planning, have
no overall emission impacts; only localized street-level NOx positive and negative changes
(+/-17%) are found due to traffic re-routing and the generation of new bottlenecks. It is
only when these measures are combined with optimistic fleet renewal as a result of the LEZ
implementation and demand reductions, that relevant global emission reductions in NOx are
obtained (-13% and -30%, respectively) with estimated NO2 reductions of -36% and -23%
at the two traffic air quality monitoring stations. Despite the potential improvements, our
simulations suggest that current measures are insufficient to comply with EU air quality
standards and that further traffic restriction policies to reduce traffic demand are needed.
Keywords: Multi-scale Air Quality Modelling, traffic simulation, Vehicle Emissions, Traffic
Management Strategies, Barcelona
Preprint submitted to Science of the Total Environment November 29, 2021
Glossary
AQD: Air Quality Directive
COPERT: Computer Program to calculate Emissions from Road Transport
CTM: Chemical Transport Model
HERMES: High-Elective Resolution Modelling Emission System
LEZ: Low Emission Zone
SPB: Superblocks
TUP: Tactical Urban Planning
VML: Virtual Mobility Lab
1. Introduction
Air pollution is a public health threat in urban areas around the world, including
Barcelona (ASPB, 2019). Although over the past years the city has been reducing its NO2
concentration levels, the traffic air quality monitoring stations still systematically exceed the
annual mean NO2 limit values established by the Air Quality Directive (AQD) 2008/50/EC
(EEA, 2019). An exception has been 2020 when no exceedances were recorded due to impact
of the COVID-19 lockdown upon mobility (ASPB, 2020). NO2 limit values are typically ex-
ceeded in traffic stations, with traffic accounting for 60% of the NO2 in the city (Ajuntament
de Barcelona, 2015). Traffic pollution is of particular concern as it has been associated with
increased mortality (Hoek et al., 2001) due to higher cancer risk (Beelen et al., 2008) and
worsened respiratory health (Brauer et al., 2002). All in all, Barcelona recently ranked as
the 6th city with higher mortality risk from a total of 858 European cities (ISGlobal (2021),
Khomenko et al. (2021)).
Partly due to the failed outcomes by the vehicle manufacturers in reducing NOx from
vehicle exhaust gases (i.e., excess diesel NOx emissions, Benavides et al. (2021a)), Barcelona
local authorities are focusing on mobility policies that try to reduce and renew the number of
circulating vehicles within the city. These policies include, on the one hand, the implemen-
tation of traffic restriction measures aiming at reducing traffic activity in certain areas and
corridors of the city and, on the other hand, the application of a Low Emission Zone (LEZ),
the objective of which is to accelerate the renewal of the circulating vehicle fleet. To reduce
traffic activity in particular urban areas, Barcelona is applying the Superblock system (Fig.
1 a,b), which consists on the traffic pacification of several streets within an area comprised by
several blocks. The traffic pacification measures performed within the superblock comprise
a reduced speed limit (10km/h), the usage of urban furniture to hassle traffic -such as urban
vegetation, bollards or speed humps- and the introduction of mandatory turnings that throw
out the incoming vehicles. The goal of the superblock is to provide a traffic-pacified interior
∗Corresponding author
Email address: daniel.rodriguez@bsc.es (Daniel Rodriguez-Rey)
2
area accessible primarily to active transport (e.g., cycling) and secondarily to residents, di-
verging through traffic and gaining space for leisure activities (Rueda, 2019). Along with the
Superblock system, Barcelona is also applying a set of Tactical Urban Planning Actions to
reduce private vehicle space of the city. These actions consist on the implementation of low-
cost and scalable elements such as strips of colours, urban furniture or moveable plant beds
to transform the urban space. In the case of Barcelona, the introduction of these elements
allowed reducing traffic lanes dedicated to private vehicles from major urban corridors in the
city, and gaining pedestrian and public transport space. The specific areas affected by SPB
and TUP actions are described in Section 2.3.1 of the manuscript and in the supplementary
material.
Figure 1: a) Area without the application of the superblock model and b) area with the superblock model.
The grey squares represent building blocks, the green segments the streets affected by traffic pacification
measures and the red segments the non-affected streets c) Image of the aplication of Tactical Urban Planning
measures in a crossroad.
To quantify the impact of these measures on emission and air quality levels, the AQD
encourages the use of numerical models. This is particularly relevant for Barcelona, where
the monitoring network is not dense enough to properly characterise population exposure,
and therefore it needs to be combined with models (Duyzer et al., 2015). Air quality models
should be used for both assessment and planning; but to do so, the modelling chain has to
be fit-for-purpose and properly validated and calibrated.
Mesoscale Eulerian Chemical Transport Models (CTMs), which require as input data
estimated emissions and meteorological variables, are currently the most widely used when
performing the evaluation of the potential emission reductions (e.g., Pisoni et al. (2019)).
These air quality systems are usually combined with traffic and travel demand models (Bar-
celó, 2011) to simulate the effect of restrictions on the traffic activity (e.g., traffic flow, speed)
across the city (e.g., Garćıa et al. (2018)). The application of such an integrated modelling
approach allows estimating the changes induced by the mobility action and their possible
rebound effects (e.g., generation of new bottlenecks) not only in the implementation area
(e.g., the LEZ), but also, more generally, in the city where it is located and its surroundings.
Despite the satisfactory performance of the CTMs, their limited resolution cannot repro-
duce the strong urban pollutant concentration gradients usually associated with high load
traffic flows (Borge et al., 2014). In order to depict street level concentration gradients,
local–scale tools are needed. The most detailed approach is the Computational Fluid Dy-
namics (CFD), which consists in the direct computation of the pollutant dispersion around
the buildings integrating the fundamental transport governing equations. In order to deal
3
with the turbulence in the CFD context, there are the Large Eddy Simulation (LES) tech-
niques that explicitly solve the larger turbulent scales while modelling the smaller ones, and
the Reynolds Average Navier Stokes (RANS) that compute the time average of each flow
characteristic. Due to the high computational load associated with CFD solvers, simpler
approaches mainly based on Gaussian dispersion models, have become very popular to assess
air quality at the urban scale, specially in large computational domains (e.g., Borge et al.
(2014), Jensen et al. (2017), Hood et al. (2018), Benavides et al. (2019)). Gaussian dis-
persion models use semi–empirical approximations to compute pollution dispersion within
the urban canopy. Local-scale tools are usually coupled with regional CTMs to account for
the long-range pollutant dispersion (Borge et al. (2018); Benavides et al. (2021a); Jensen
et al. (2017); Hood et al. (2018); Denby et al. (2020)). Besides improving the performance
of grid-based CTMs, the application of a hybrid air quality modelling approach (i.e. the
combination of mesoscale and street-scale dispersion models) can also lead to more detailed
results when evaluating urban mobility plans.
The coupling of such hybrid approach with traffic demand models has already been ap-
plied to estimate the impact in urban emissions and air quality levels of different structural
or short term action plans. Borge et al. (2018) estimated the impact of the Madrid NO2
protocol using a multi-scale air quality model with a mesoscale approach (1 × 1 km2) for the
citywide, and a street-level approach for one of the main streets in the city. Using different
estimated traffic demand scenarios, their results showed that only the most restrictive mea-
sure would produce a noticeable air quality improvement (-25% in NOx emissions). Jensen
et al. (2017) estimated the NO2 concentration values in Denmark using a multi-scale air
quality modelling approach. In this case, they combined the Danish Transport Model and
GPS vehicle readings to estimate traffic volume and speed, respectively. This data was used
by AirGis (Jensen et al., 2001) to estimate NO2 concentration values at different resolutions
with values that ranged between -27% and +12% from observations. In Barcelona, Mueller
et al. (2020) studied the health impacts of the complete implementation of the Superblock
idea on the city (500 superblocks). They used the Street 5.2 air quality model (Kunz, 2005)
estimating a reduction of 24% in the NO2 concentration values assuming a 19% reduction in
the private transport. They estimated 291 premature deaths due to NO2 following a linear
exposure-response function (Atkinson et al., 2018). Since the study followed a macroscopic
approach obtaining the results at the city level, the most affected areas or streets as well as
possible rebound effects were not identified.
Our study quantifies the air quality changes at street level of the different traffic manage-
ment strategies adopted in Barcelona and for the first time their potential undesired rebound
effects as a consequence of the new generated vehicle routes. To address this, we perform
a multi-scale air quality modelling exercise that couples the CALIOPE-Urban, a modelling
framework that combines the CALIOPE mesoscale air quality system (Baldasano et al.
(2011); Pay et al. (2014b)) with the street-scale dispersion model adapted to street canyons
(Benavides et al., 2019) and the traffic-emission model VML – HERMESv3 (Rodriguez-Rey
et al., 2021) that allows to estimate the induced traffic emissions as a consequence of road
network modifications. Additionally, we compare the mesoscale and street scale computed
values to understand and quantify the differences when working at different scales.
4
2. Methodology
In this section we present the domain and period of study, as well as a detailed description
of the model workflow. We also describe the traffic restrictions considered in the study along
with their implementation in the modelling system.
2.1. Selection of the domain and period of study
The area of study is the greater area of Barcelona, which covers a surface of 101km2
and is home to about two million inhabitants (Fig. 3a) holding one of the highest traffic
densities in Europe (6000 vehicles/km2 (Ajuntament de Barcelona, 2020a)). The study is
performed from the 9th to the 25th of November of 2017. This period includes representative
days of the observed NO2 annual mean daily cycle of the city (9th - 16th) and a pollution
episode with high NO2 concentration levels (17th - 25th) to assess the impact of the traffic
restriction measures in a critical air quality condition in the city. During this period, there
was an exceedance of the NO2 hourly limit value (200 µg/m3), and six values above 160
µg/m3 of NO2 at the two traffic air quality monitoring stations and in a urban background
site.
The NO2 mean for the simulated study period and the NO2 mean for the observed
annual values for Eixample are of 63 (+/- 30) µg/m3 and of 58 (+/- 26) µg/m3, respectively.
The simulated study period NO2 mean for Gràcia is of 60 (+/- 31) µg/m3, while the annual
observed NO2 mean is of 51 (+/- 28) µg/m3. A detailed description of the representativeness
of the selected study period with the annual mean daily cycle can be found in (Benavides
et al., 2021b).
2.2. Modelling system
The emission and air quality impacts of the different traffic management strategies ap-
plied in Barcelona were analyzed using a multi-scale modelling chain based on the coupled
traffic-emission model VML-HERMESv3 (Rodriguez-Rey et al., 2021), the mesoscale air
quality system CALIOPE (Baldasano et al. (2011), Pay et al. (2014b)) and the street scale
air quality system CALIOPE-Urban (Benavides et al., 2019) (Fig. 2).
2.2.1. Traffic and emission modelling: VML-HERMESv3
Traffic flow and speed are simulated using the Barcelona Virtual Mobility Lab (VML)
model (Montero et al., 2018). The VML, which relies on the VISUM traffic simulator (PTV
Group, 2019), contains information for each network link of the maximum allowed speeds,
lane capacities, permitted turns, and other traffic constraints that determine the traffic
assignment (e.g., vehicle flow and speed) for the study domain area. Vehicle routing is
estimated according to mobility matrices (e.g., Origin-Destination matrices) derived from
mobile phone data from March 2017. An Static Traffic Assignment is performed for each
hour for a business-as-usual day (for a total of 24 traffic simulations). A calibration process
using measured traffic counts and speed values from 138 permanent traffic stations was per-
formed to ensure correct values of traffic flow and speed. Following the traffic simulation, the
averaged link-level hourly speed, traffic flow data and the traffic network from the VML are
5
Figure 2: Schematic representation of the multi-scale modelling chain used to evaluate the impact of traffic
management strategies in Barcelona. Blue dotted lines comprise the modelling systems that compose the
tool (VML-HERMESv3: coupled traffic and emission model, CALIOPE: mesoscale air quality system and
CALIOPE-Urban: urban scale air quality system), black boxes represent input data, grey boxes stand for
the inner models, and black arrows illustrate the transfer of data between models and systems.
read by the HERMESv3 traffic emission model to compute vehicle emissions at the road link
level. The coupling process is script-based and assures the correct link mapping, its assigned
vehicle composition and the speed and flow temporal profiles. HERMESv3 traffic (Guevara
et al., 2020) computes hourly exhaust (hot and cold-start) and non-exhaust emissions (tire,
road and brake wear and resuspension) based on the Computer Program to calculate Emis-
sions from Road Transport version V (COPERT V, which includes the corrected Diesel
emission factors) (EMISIA, 2016), which are included in the tier 3 approach of the Euro-
pean emission inventory guidelines EMEP/EEA (EMEP/EEA, 2017). To compute vehicle
emissions, HERMESv3 traffic uses the link-level average speed, traffic flow, temporal pro-
files and the vehicle fleet composition previously assigned to each link. A total of 5 different
fleet composition profiles assigned to different regions of the city are considered. These are
derived from a remote-sensing device campaign performed in 2017 (AMB and RACC, 2017),
and information provided by the Barcelona’s Port Authority (Port de Barcelona, 2017). A
detailed description of the coupling and calibration of the VML and the HERMESv3 traffic
models can be found in Rodriguez-Rey et al. (2021).
2.2.2. Mesoscale air quality modelling: CALIOPE
The mesoscale air quality modelling system CALIOPE is composed of the Weather Re-
search and Forecasting (WRF) model version 3.5.1 with WPS (WRF Preprocessing System)
6
version 3.9.1 (Skamarock and Klemp, 2008), combined with the HERMESv3 multi-scale at-
mospheric emission modelling framework (Guevara et al., 2019, 2020) and the Community
Multiscale Air Quality Modelling System version 5.0.2 (CMAQ) (Byun and Schere, 2005).
To obtain the adequate boundary and initial conditions, the CALIOPE system is run over
Europe at a 12 km by 12 km horizontal resolution, the Iberian Peninsula at 4 km by 4 km,
and the Catalonian region including Barcelona at 1 km by 1 km resolution (Fig. 2). A
one-way nesting is performed to retrieve the meteorological and chemical conditions from
one domain to the inner ones. An evaluation of CALIOPE results can be found in Baldasano
et al. (2011), Pay et al. (2014a), and Guevara et al. (2014). CALIOPE simulations were
initialised with the Global Forecast System (GFS) (NCEP, 2011), boundary conditions for
chemistry come from the CAMS reanalysis of atmospheric composition (Inness et al., 2019),
and anthropogenic emissions for all domains were processed using HERMESv3. For the Eu-
ropean domain, HERMESv3 was run using the TNO-MACCIII (Kuenen et al., 2014) and
the HTAPv2.2 (Janssens-Maenhout et al., 2015) inventories for European and non-European
countries, respectively. For the Iberian Peninsula and Catalonia nested domains, emissions
were estimated using the bottom-up module of HERMESv3, except for Barcelona domain
where the estimated traffic emissions were computed by the VML-HERMESv3 coupled sys-
tem previously described. The bottom-up module of HERMESv3 includes source-level emis-
sion estimation approaches mostly based on the Tier 3 calculation methodologies reported
by the European EMEP/EEA air pollutant emission inventory guidebook. Further infor-
mation on the methods and datasets used by HERMESv3 to compute bottom-up emissions
can be found in Guevara et al. (2020). Biogenic emissions were computed for all the three
domains using the MEGANv2.0.4 model (Guenther et al., 2006).
2.2.3. Street-scale air quality modelling: CALIOPE-Urban
The CALIOPE-Urban street-scale air quality model is applied over the Barcelona domain
(Fig. 2) at a resolution of 20 meters. This resolution is selected to account for the strong
pollutant concentration decay within tens of meters from the road edge (e.g. Black Carbon
decay of more than 50% in street canyons within the initial 20 m (Amato et al., 2019))
that occur in compact cities while keeping feasible computational time and memory require-
ments. CALIOPE-Urban is composed of the above-mentioned CALIOPE air quality system
- from where it takes the meteorology and chemical boundary conditions - and the near
road dispersion model R-LINE (Snyder et al., 2013) adapted to street canyons (Benavides
et al., 2019). The meteorology and chemical boundary conditions modelled by CALIOPE
are downscaled to street level following the parametrisations described in Benavides et al.
(2019). CALIOPE-Urban allows to estimate local traffic dispersion driven by channelled
street winds and vertical mixing considering background NO2 and O3 concentrations, atmo-
spheric stability, and street morphology. R-LINE applies the Generic Reaction Set (GRS) to
resolve simple NO to NO2 chemistry (Valencia et al., 2018), whose major limitation of the
chemistry computations at the urban scale is the assumption of clear-sky conditions, which
neglects the cloud-effects on the NO2 photolysis rate. The street-scale dispersion model is
run using the street-scale and hourly traffic emissions estimated by the VML-HERMESv3
system. These emission values are consistently used for both the street-scale (BCN-20m)
7
and the mesoscale (CATALONIA-1km) modelling domains. Further information regarding
CALIOPE-Urban can be found in Benavides et al. (2019).
2.3. Traffic restriction measures
We consider 3 different traffic restriction measures that are currently being applied or
planned to be applied in the near future in Barcelona: (I) Tactical Urban Planning -TUP-
, (II) Superblocks -SPB- and (III) the Low Emission Zone -LEZ-. The LEZ is applied
in almost all the simulated domain, with some excluded areas and ways (Fig. 3a). It
forbids the entrance of gasoline vehicles below Euro 3, diesel vehicles below Euro 4 and
motorbikes/mopeds below Euro 2. Detailed descriptions of these measures can be found in
Area Metropolitana de Barcelona (2020) and Ajuntament de Barcelona (2021b). Based on
the set of measures presented by the Barcelona City Hall, we consider TUP actions that
aim to remove vehicle lanes in major urban corridors of the city. We have reduced a total of
31.67 km from vehicle lanes, represented by the dotted blue lines in Fig. 3b. Additionally,
we have implemented a total of eight superblock areas represented by the orange polygons
in Fig. 3b. The considered superblocks are the ones with specific information of the planned
actions made within. The specific streets modified and superblocks applied can be found in
the supplementary material.
Figure 3: Barcelona study domain showing a) the area where the Low Emission Zone is applied (orange)
and b) the superblocks (orange squares), the streets where tactical urban planning (TUP) actions were
implemented (dashed blue lines), and the Barcelona urban traffic air quality monitoring stations (yellow
dots). Adapted from Ajuntament de Barcelona (2020d). The ortogonal image from a) was obtained from
Leaflet — Tiles © Esri — Source: Esri, i-cubed, USDA, USGS, AEX, GeoEye, Getmapping, Aerogrid,
IGN, IGP, UPR-EGP, and the GIS User Community.
In order to quantify the impact of each of these measures and differentiate their specific
potential undesired rebound effects, we have simulated them individually and collectively,
as specified in Table 1. On top of these scenarios, we have also modelled a Base Case
8
(business-as-usual) scenario, which is taken as the reference for the comparison. The criteria
to select the simulated scenarios was based on the ability of the traffic-emission-air quality
system to estimate the new vehicle route equilibrium and the traffic associated emissions.
Hence, the unique application of measures affecting vehicle emission factors (i.e., LEZ) were
not considered for air quality simulations since similar studies have already been published
elsewhere (Soret et al., 2014).
Table 1: List of scenarios and combination of traffic management measures considered in the study (TUP,
Tactical Urban Planing; SPB, Superblocks; and LEZ, Low Emission Zone). Emissions have been modelled
for all the scenarios, while we specify the scenarios where the air quality impact (AQ) has been modelled.
nº Scenario AQ
1 Base Case: Original 2017 network Yes
2 Base Case + TUP No
3 Base Case + SPB No
4 Base Case + TUP + SPB Yes
5 Base Case + LEZ No
6 Base Case + TUP + SPB + LEZ Yes
7 Base Case + TUP + SPB + LEZ + demand reduction (-25%) Yes
One of the major difficulties in simulating the effects of traffic restrictions on vehicle
fluxes is to estimate how the demand (e.g., total number of trips) will be affected due to
their implementation. In this study, we assumed both the most pessimistic and optimistic
scenarios, which are: (i) the number of total circulating vehicles will not change despite
the measures implemented; and (ii) the traffic demand will suffer a -25% reduction once
all the three different strategies are fully implemented, which is the expected reduction
estimated by the Barcelona city council (Ajuntament de Barcelona, 2021b). In total, we
simulated seven NOx emission scenarios (one for each combination of traffic management
strategies) and modelled the impact on NO2 air quality levels in four of them due to the large
computational load that it requires (Table 1). NO + NOx (NOx) and VOCs were simulated
for all emission scenarios although only NOx is represented since VOCs are used for the GRS
chemical processes of CALIOPE-Urban. Even though Particular Matter (PM) is a pollutant
with severe health issues, it will not be simulated in this study since the GRS module used
by R-Line can only compute the dispersion of primary species (e.g., black carbon).
2.3.1. Tactical Urban Planing and Superblocks scenario
Changes on the traffic flow and speed induced by the TUP and SPB strategies were mod-
elled following the description of measures reported in Ajuntament de Barcelona (2020b,c).
To do so, we have modified target features of the original VML road network. The mod-
ifications include the removal of traffic lanes, changes in the street capacities and allowed
maximum speed, and the addition of new turns to characterize SPB measures. The reduc-
tion of vehicle lanes of the TUP measures was directly applied to the VML network. The
VML road capacity is calculated by the multiplication of the user-defined number of lanes
9
and capacity of each lane, which depends on the specific road type (e.g., highway, urban
corridor, etc). The number of lanes of the street links affected by the TUP measures were
manually modified. Specific information of the modified links can be found in the supple-
mentary material. For the SPB areas, in addition to the above-mentioned modifications, the
construction of traffic-pacified street segments was modelled using a new segment type with
a specific reduced capacity and speed. All modifications are made to prevent the circulation
of modelled traffic flow as shown in Fig. 4.
Figure 4: Screenshot of the VML traffic simulator network for (a) the Base Case scenario, and for (b)
scenario SPB where a specific road type has been generated (blue dotted line) to represent the superblock
restrictions. The colored segments indicate other street types with different parameters restricting vehicle
movement (e.g., as a consequence of a traffic light green wave). From higher movement to lower: Red,
yellow, grey, dottet blue. Detailed information about the traffic simulation can be found in Rodriguez-Rey
et al. (2021).
2.3.2. Low Emission Zone scenario
The LEZ scenario, which forbids the entrance of the most polluting vehicles in the city,
was simulated by adapting the description of the vehicle fleet composition profiles used
in HERMESv3 to the restrictions associated to this measure. We assumed an optimistic
scenario in which all banned vehicles are replaced by new Euro 6 vehicles. We wanted
to reflect the tendency of drivers to switch from diesel to gasoline cars that can already
be detected from the sales of gasoline-powered cars (European Automobile Manufacturers
Association, 2021). To do so, we used the new registered vehicle data in Barcelona for 2019
(DGT, 2020), and applied the observed gasoline/diesel sales distribution to the new Euro
6 vehicles that appeared as a consequence of the LEZ. We did not substitute the banned
vehicles by zero or low emission vehicles based on the Barcelona city hall expected vehicle
fleet composition renewal for the next years, which accounts for less than 3% of zero or low
emission vehicles by 2022 (Institut Cerdà, 2019).
Figure 5 compares the observed vehicle fleet composition in 2017 (AMB and RACC,
2017) and the estimated one after applying the LEZ. The share of Euro 6 diesel vehicles
increases from 14% to 20%, while the share of Euro 6 gasoline vehicles increases from 6%
10
to 19%. The obtained LEZ fleet composition was applied to the road links within the LEZ
domain represented by the green area in Fig. 3a. Due to the difficulties in assessing the
LEZ impact on traffic flow (Holman et al., 2015), this scenario does not imply reductions in
the simulated traffic demand.
Figure 5: Observed vehicle fleet composition in 2017 (blue) and estimated Low Emission Zone (LEZ) vehicle
fleet composition (orange) per vehicle type and Euro category. PC refers to Passenger Car, HDV to Heavy
Duty Vehicle, and LDV to Light Duty Vehicle.
3. Results and discussion
Section 3.1 presents the NOx emission results computed for the Base Case scenario and
a comparison of the CALIOPE and CALIOPE-Urban NO2 modelled concentrations against
observations registered at the two urban traffic monitoring stations shown in Fig. 3b. The
model performance at the other urban background monitoring stations can be found in
the supplementary material. The emissions and air quality results for each of the traffic
management scenarios are presented and discussed in Section 3.2 and 3.3 respectively.
3.1. Base Case scenario
Figures 6a and 6b show the road traffic emissions for the Base Case scenario used as
input in the CALIOPE (1 km × 1 km) and CALIOPE-Urban (20 m × 20 m) systems,
respectively. The VML-HERMESv3 estimated link-level vehicle emissions which are used
by CALIOPE-Urban, are mapped onto the 1 × 1 km2 gridded domain for CALIOPE in
order to ensure consistency between modelling scales. Note that in Fig. 6b emission results
are presented at a 50 m × 50 m resolution only for representation purposes. Both figures
show that the highest NOx values are found in the ring roads and the city center, where
the main urban corridors are found. However, the street-level results allow representing the
strong emission gradients that cannot be reproduced at 1 km × 1 km. Similarly, a large
heterogeneity of the NO2 distribution is observed with CALIOPE-Urban (BCN-20m), while
the mesoscale CALIOPE (BCN-1km) modelling results show a rather uniform distribution
of NO2 levels across the city, the largest values found near the port area (Fig. 6c and Fig. 6d).
11
Figure 6: Average daily NOx traffic emissions (kg/day) modelled by VML-HERMESv3 at a resolution of 1
km × 1km (a) and 50 m × 50 m (b), and average daily NO2 concentrations (µg/m3) modelled by CALIOPE
and CALIOPE-Urban at a spatial resolution of 1 km × 1 km (c) and 20 m × 20 m (d).
Figure 7 shows the NO2 average diurnal cycle observed at the Eixample and Gràcia sta-
tions (black dotted lines) and modelled by the CALIOPE and CALIOPE-Urban systems for
the Base Case scenario (brown and pink lines, respectively). Standard statistics computed
using the modelled and simulated concentrations are summarised in Table 2. CALIOPE-
Urban is capable of capturing the morning and afternoon NO2 peaks at the Eixample station,
while it slightly overestimates the afternoon peak hour at Gràcia by 3% and estimates it
earlier. Yet, the system is not able to properly capture the concentrations observed dur-
ing the valley hours (11-16h), and it presents an underestimation of 30 µg/m3 (-38%) in
both stations. This behaviour in CALIOPE-Urban is inherited from the CALIOPE regional
model, which cannot properly reproduce the vertical mixing of pollutants during daytime.
Results clearly indicate that CALIOPE-Urban improves the modelling performance in cap-
12
turing the NO2 peaks with respect to the mesoscale approach (CALIOPE) while providing
a more detailed information at the street-level. In both stations CALIOPE-urban reduces
the mean bias observed with CALIOPE by half (-12 and -16 µg/m3 vs -24 and -33 µg/m3,
respectively) and increases the FAC2 (fraction of predictions within a factor of two obser-
vations) by 0.21 and 0.11 in Gràcia and Eixample respectively, while maintaining a similar
correlation (0.54 vs 0.58 and 0.52 vs 0.55 for Gràcia and Eixample).
Table 2: NO2 model evaluation statistics calculated for the urban (CALIOPE-Urban) and mesoscale
(CALIOPE) system at Eixample and Gràcia stations for the period of study (7 to 25 November 2017).
FAC2 refers to the fraction of modelled results within a factor of 2 of observations, MB refers to the mean
bias, RMSE refers to root-mean-square Error, and r to the correlation coefficient.
Site Model FAC2 MB RMSE r
CALIOPE 0.67 -23.99 38.01 0.52
Eixample
CALIOPE-Urban 0.77 -12.00 32.65 0.55
CALIOPE 0.54 -33.34 46.36 0.54
Gràcia
CALIOPE-Urban 0.75 -15.88 36.00 0.58
Figure 7: NO2 average diurnal cycle observed (dotted back) and modelled with CALIOPE-Urban for the Base
Case scenario (brown), scenario 4 (green), scenario 6 (sky blue) and scenario 7 (blue), and with CALIOPE
for the Base Case scenario (pink) at Gràcia and Eixample urban traffic stations during the period of study
(7 to 25 November 2017).
3.2. Impact of traffic strategies on NOx emissions
Figure 8 shows the aggregated NOx daily mean emissions for the base case scenario over
the complete modelling domain, and the relative difference with the scenarios 2 to 7 for the
period of study. According to these results, emissions are only reduced on those scenarios
where the LEZ or the traffic demand reduction of -25% are implemented (scenarios 5 to 7).
For all the other scenarios, when only TUP or SPB strategies are considered, the impact on
the total NOx emissions is negligible (+0.1%, due to the new generated congested areas).
13
Maximum reductions ( -30%) are achieved when all the traffic management strategies are
combined and a -25% reduction of the number of vehicles is assumed. The combination
of all strategies without changes on the number of circulating vehicles shows overall NOx
emission reductions of -13.1%.
Figure 8: NOx daily mean emissions (kg) for all scenarios and the relative differences (%) between the Base
Case and the other simulated scenarios.
Despite not having any impact on the overall emissions, a significant redistribution of
emissions at the street level can be observed when only implementing the TUP and SPB
measures (scenarios 2 - TUP -, 3 - SPB -, and 4 - TUP+SPB -), which is caused by the re-
routing of traffic assignment. To illustrate that, Fig. 9 shows the daily mean NOx difference
in an aggregated grid of 100 m × 100 m between the scenarios 2 to 7 with the Base Case
scenario. Note that emissions at Fig. 9 are shown in a grid of 100m for representation
purposes, but the original emission outputs are obtained and passed to CALIOPE and
CALIOPE-Urban at the road link level. Since scenario 4 is the one combining TUP and
SPB measures, in Fig. 10 we take a closer look into how the traffic flow (in annual average
daily traffic, AADT) and associated NOx emissions change in five representative street cells.
The mentioned streets over the analysed emission cells are illustrative, since the 100 m ×
100 m emission grid cell could also contain parts of other street segments. In this sense, the
cell in one of the urban corridors with higher traffic load and emissions, Aragó (1) (Fig. 10),
shows an average reduction of -17% (-1.3 kg NOx/100m2) in the simulated NOx emission
levels along the cell with a traffic flow reduction of -24% (-21.000 vehicles/day) (Fig. 10). On
the other hand, NOx emission increases up to +17% (+0.5 kg NO /100m2
x ) are observed in
other adjacent street cells (e.g., Tarragona (4) or Viladomat (5)) as a consequence of a traffic
flow increase of +30% (+6.000 vehicles/day) and +125% (+5.500 vehicles/day) respectively
(Fig. 10). Viladomat (5) is of particular interest since the street is within a superblock area
but it is still absorbing the traffic through it. In the view of these values, we have not found
evidence of linearity between the absolute or relative traffic flow increment/decrement with
the absolute or relative increase or decrease in NOx emissions. This might be caused by the
emission dependency in both traffic flow and speed, by the presence of other segments within
the grid adding extra emissions and the difficulties of the macroscopic system in recreating
vehicle emissions under congested situations.
14
Figure 9: Daily mean NOx emission differences (kg/100m2) from the Base Case scenario with the scenarios
2 to 7 for the period of study. Scenario 4 also shows the location of the streets commented in Fig 10: Aragó
(1), Diagonal (2), Gran Via (3), Tarragona (4) and Viladomat (5).
15
Figure 10: Annual average daily traffic (AADT) and daily mean NOx emissons (kg) of the streets mentioned
in Fig. 9 for the scenarios Base Case, 4, 6 and 7.
As described in Rodriguez-Rey et al. (2021), the macroscopic system cannot properly
estimate the location of bottlenecks and the start-stop vehicle behaviour, which implies
an underestimation of vehicle emissions in these situations. Our results reveal that large
superblocks cannot avoid the through pass of vehicles within the designated area if traffic
pacification measures are not applied in all streets within the superblock (Fig. 4). Smaller
superblocks like the Poble Nou and Sarrià show NOx emission decreases in all their streets
that can reach up to -15%. This is due to the already low traffic load of such areas, that
now discourage the through traffic, becoming areas used only by residents and services.
3.3. Impact of traffic strategies on NO2 air quality levels
Figure 11 illustrates the differences between modelled NO2 concentrations averaged over
the period of study in scenarios 4, 6 and 7 and the Base Case scenarios using CALIOPE
(left column) and CALIOPE-Urban (right column) systems. In line with the emission re-
sults, the impact of traffic measures on NO2 concentrations is mainly observed when LEZ
and/or the traffic demand reduction of -25% are implemented (scenarios 6 and 7). Under
these two scenarios (6 and 7), NO2 daily mean reductions of -5 to -10 µg/m3 (-10% to -
20%) and of -10 to -25 µg/m3 (-20% to -30%) are observed in the main urban corridors,
respectively, during the period of study. In both scenarios, the largest NO2 decreases oc-
cur in the most trafficked streets due to the presence of less polluting vehicles (LEZ) and
the implementation of TUP and SPB measures in such streets. When only TUP and SPB
measures are considered (scenario 4), NO2 impacts are very limited. NO2 reductions and
increases of similar intensity (+/-5 µg/m3) are observed as a function of the street due to
the traffic re-routing, as explained in section 3.2. The strong street gradients and associated
NO2 concentration changes modelled with CALIOPE-Urban cannot be reproduced with the
16
CALIOPE mesoscale system, which captures NO2 reductions of only -6 to -10 µg/m3 and
between -3 to -6 µg/m3 for the scenarios 7 and 6, respectively.
As shown in Table 3, with CALIOPE-Urban and under the most restrictive scenario
with a demand reduction of -25% (scenario 7) the NO2 averaged daily maximums would be
reduced by a -36% and -22.5% in Gràcia and Eixample monitoring stations. This would
represent a reduction of the observed daily maximum from 108 to 69 µg/m3 (-39 µg/m3)
and from 101 to 78 µg/m3 (-23 µg/m3) of NO2 for Gràcia and Eixample monitoring stations.
If we perform the same exercise for the average daily mean, the observed NO2 daily mean
during the study period for Gràcia would be of 45.5 µg/m3 (-38%) and of 52.4 µg/m3 (-
26.6%) for Eixample. On the other hand, the NO2 reductions captured by the mesoscale
CALIOPE system are around half (-18% and -17% for the daily average mean and daily
maximum, respectively).
Table 3: Period-averaged daily mean and period-averaged daily maximum modelled NO2 values (µg/m3) for
the Base Case scenario and the relative difference with the scenarios 4, 6 and 7 at the Gràcia and Eixample
air quality station.
CALIOPE-Urban CALIOPE
Daily Daily Daily Daily
Scenario Station
mean average maximum mean average maximum
NO2 base case Gràcia 57.4 111.0 39.3 82.0
(µg/m3) Eixample 59.2 105.0 46.5 87.0
Scenario 4 Gràcia -18.6% -18.7% 0.0% 0.0%
(Diff. %) Eixample -4.4% -2.0% 0.0% 0.0%
Scenario 6 Gràcia -28.6% -26.2% -10.4% -6.1%
(Diff. %) Eixample -15.0% -11.8% -10.8% -8.0%
Scenario 7 Gràcia -38.0% -36.4% -18.1% -17.1%
(Diff. %) Eixample -26.6% -22.5% -18.3% -17.2%
Our work provides an estimate of the expected emissions and concentration levels after
the application of the traffic restrictions in Barcelona. Based on the relative reductions
obtained, we estimate that all the simulated scenarios would fail in reducing the annual
average value below the AQD limit of ≤ 40 µg/m3 of NO2 in the two traffic stations for the
considered study period. Yet, under scenarios 6 and 7, the NO2 hourly exceedance of 216
µg/m3 that occurred in November 2017 would very likely have been avoided.
17
Figure 11: Daily mean NO2 concentration difference (µg/m3) between the base case and the scenarios 4, 6
and 7 for the November period. Left column shows the mesoscale values from CALIOPE model (1km) while
right column shows the CALIOPE-Urban street scale model results (20m). Bluewish and yellowish colors
indicate a decrease and increase in emissions, respectively.
18
4. Conclusions
We quantified the impact of different traffic management strategies on the NOx emissions
and NO2 concentration levels in the city of Barcelona. To do so, we performed a multi-scale
air quality modelling exercise that combines mesoscale (1×1 km) and street-scale (20×20
meters) air quality modelling tools, fed by a coupled macroscopic traffic and vehicle emission
model. A total of 7 NOx emission scenarios and 4 NO2 air quality scenarios were simulated
during a bi-weekly time period, which coincides with an air pollution episode that took
place in November 2017. The scenarios comprise simulations of individual and varying
combinations of the mobility measures currently being applied in Barcelona: a Low Emission
Zone (LEZ), which aims to forbid the entrance of most polluting vehicles, Tactical Urban
Planing (TUP) measures, which reduce the space that private transport has on specific
streets of the city, and the Superblocks (SPB), which prevent the entrance of vehicles in a
certain urban area delimited by several blocks. We additionally added a scenario where all
these measures were combined with a traffic demand reduction of -25% based on the City
Hall’s expectations over the next years. The main findings and conclusions of this work are
as follows:
− The largest road transport NOx emission reductions in Barcelona (-30%) are obtained
when combining all traffic management strategies (LEZ, TUP and SPB) together with
the -25% traffic demand reduction. The computed reductions are only of -13% when
traffic demand is kept constant.
− When only implementing measures related to the reduction of private transport space
(TUP and SPB), overall NOx changes are negligible (+0.1%). Nevertheless, these
measures generate a traffic redistribution along the network, which shows a noticeable
impact on the street-level emission distribution, with NOx variations up to +/-17% in
specific streets as a consequence of the new vehicle routing and the variation in traffic
flow and speed. We have not found evidence of linearity between the traffic flow and
speed variation and the increment or decrement in NOx emissions.
− The impact on NO2 air quality levels follows the same pattern as for emissions. Scenar-
ios comprising all the traffic strategies with and without the -25% demand reduction
show the largest NO2 reductions at the main urban corridors (with daily mean reduc-
tions of -10 to -25 µg/m3 and -5 to -10 µg/m3, respectively and daily max reductions
up to -70 µg/m3 and -50 µg/m3, respectively). When only implementing the TUP and
SPB measures, limited increases and decreases of NO2 are observed (+/-5 µg/m3) due
to traffic re-routing.
− The expected differences at the two traffic air quality monitoring stations of the city
(Gràcia and Eixample) when implementing all traffic strategies and the demand reduc-
tion are of -38% and -27% for the daily mean average, and of -36% and -23% when con-
sidering the averaged daily max values for Gràcia and Eixample, respectively. These
values decrease by approximately a 10% without considering the demand reduction.
19
− The NO2 daily mean and peak concentration reductions modelled with the mesoscale
system are almost two times lower than the ones obtained with the urban scale system.
Despite presenting consistent results, the mesoscale system is not capable of modelling
the strong street gradients and associated NO2 concentration changes induced by the
mobility restrictions. Considering these results, we recommend that the evaluation
of urban traffic management strategies is performed applying a multi-scale modelling
approach with street scale resolution for an optimal result.
− Although traffic management strategies lead to significant emissions and air quality
improvements, the reductions achieved are insufficient to ensure proper air quality
levels. These strategies must be accompanied by a large decrease in the total number
of vehicles circulating throughout the city.
The methodology followed in this study can provide highly-detailed data for health as-
sessment studies, like the one performed by Mueller et al. (2020). The authors are aware of
the important health-related implications of Particular Matter and we are working towards
including this pollutant in future works of this modelling system. On the same direction, the
developed tool could be used for an ex-ante evaluation of further mobility policies applied
by policy-makers, providing some insights before its application. In Barcelona, since the
current measures appear to be insufficient to reduce the NO2 levels below the EU AQD,
the authors suggest the study of other proven-effective measures in reducing traffic demand
like the congestion charging scheme, with successful results in Milan, London or Stockholm
(Croci, 2016) combined with a more restrictive LEZ (Bigazzi and Rouleau, 2017).
This research shows that the application of a multi-scale air quality tool, combining
traffic, emission and street dispersion models is essential to properly capture the induced NOx
and NO2 variations generated by the application of different traffic management strategies.
We acknowledge the limitations of the macroscopic traffic model to properly recreate the
start-stop vehicle behaviour under congested situations, which implies an underestimation in
vehicle emissions, as shown in Rodriguez-Rey et al. (2021). The macroscopic tool is however
appropriate to estimate the impact of the analysed strategies at urban scale. The usage of
a microscopic tool able to properly estimate traffic flow and vehicle emissions at the studied
domain it is currently not a viable option due to the data and computational load needed.
We also acknowledge the limitations of the study related to traffic demand modelling. It
was assumed constant over the different scenarios studied, with the exception of scenario 7
which had a demand reduction of -25%. In reality, a modal transfer could occur between the
private and public transport, or to other active ways of mobility (e.g., cycling) due to the
private transport restrictions. In future works we plan to keep with the analysis of possible
new TMS such as the application of green corridors (Ajuntament de Barcelona, 2021a) or
to study the impact of additional measures that are currently not being considered by the
city council such as the implementation of a congestion charge.
20
Author contributions. DRR, MG, MPL, JC and AS designed the research. JB and JMA
participated in the building and calibration of the CALIOPE-Urban model. DRR performed
the model coupling process and ran the experiments assisted by MG, JMA, OJ, JB and CT.
MG, CP and OJ supervised the work. DRR prepared the paper with contributions from all
co-authors.
Interests statement. The authors declare that they have no conflict of interest
Acknowledgements. The authors acknowledge CARNET-The Future Mobility Research
HUB to allow the usage and work on the BCN-VML network, as well as PTV VISUM
for the traffic software license. The author thankfully acknowledges the computer resources
at MareNostrum and the technical support provided by Barcelona Supercomputing Center
(PRACE, AECT-2020-1-0007, AECT-2021-1-0027). The authors also acknowledge the sup-
port from the Ministerio de Ciencia, Innovación y Universidades (MICINN) as part of the
BROWNING project RTI2018-099894-B-I00 and NUTRIENT project CGL2017-88911-R
and the support from the Agencia Estatal de Investigacion (AEI) as part of the VITALISE
project (PID2019-108086RA-I00 / AEI / 10.13039/501100011033). Carlos Pérez Garćıa-
Pando acknowledges the long-term support from the AXA Chair in Sand and Dust Storms
(AXA Research Fund), as well as the support received through the Ramón y Cajal pro-
gramme (grant RYC-2015-18690) of the Spanish Ministry of Economy and Competitiveness.
Funding This research has been funded by the grant BES-2016-078116 from the FPI pro-
gram by the Spanish Ministry of Economy and Competitivenes as well as the VITALISE
project (PID2019-108086RA-I00 / AEI / 10.13039/501100011033) from the Agencia Estatal
de Investigacion (AEI).
21
References
Ajuntament de Barcelona. Plà de millora de la qualitat de l’aire a Barcelona 2015-2018. Technical report,
2015.
Ajuntament de Barcelona. Sabies que Barcelona es la ciutat d’Europa amb més densitat de vehicles?, 2020a.
URL https://ajuntament.barcelona.cat/qualitataire/ca/noticia/
sabies-que-barcelona-es-la-ciutat-deuropa-amb-mes-densitat-de-vehicles 403904.
Ajuntament de Barcelona. Una Mobilitat sostenible en un nou Espai Pùblic. Technical report, 2020b.
Ajuntament de Barcelona. Una Mobilitat sostenible en un nou Espai Pùblic - Fase 2. Technical report,
2020c.
Ajuntament de Barcelona. Superilla Barcelona, 2020d. URL
https://ajuntament.barcelona.cat/superilles/sites/default/files/
Presentacio SUPERILLA BARCELONA.pdf.
Ajuntament de Barcelona. Superilla Barcelona: Guanyadors dels concursos eixos verds i places, 2021a. URL
https://ajuntament.barcelona.cat/superilles/sites/default/files/Guanyadors concurs SuperillaBarcelona.pdf.
Ajuntament de Barcelona. Pla de mobilitat urbana 2024. Technical report, 2021b. URL
https://www.barcelona.cat/mobilitat/sites/default/files/documentacio/
pmu bcn 2024 per ceuim 20201214 compressed.pdf.
F. Amato, N. Pérez, M. López, A. Ripoll, A. Alastuey, M. Pandolfi, A. Karanasiou, A. Salmatoni-
dis, E. Padoan, D. Frasca, M. Marcoccia, M. Viana, T. Moreno, C. Reche, V. Martins, M. Brines,
M. Minguillón, M. Ealo, I. Rivas, B. van Drooge, J. Benavides, J. Craviotto, and X. Querol. Vertical
and horizontal fall-off of black carbon and no2 within urban blocks. Science of The Total Environ-
ment, 686:236–245, 2019. ISSN 0048-9697. doi: https://doi.org/10.1016/j.scitotenv.2019.05.434. URL
https://www.sciencedirect.com/science/article/pii/S004896971932488X.
AMB and RACC. Caracterització dels vehicles i les seves emissions a l’àrea metropolitana
de Barcelona. Technical report, Àrea Metropolitana de Barcelona, Barcelona, 2017. URL
http://saladepremsa.racc.cat/wp-content/uploads/2017/09/
Dossier estudi Vehicles-i-Emissions-Area-Metropolitana.pdf.
Area Metropolitana de Barcelona. Zona de Baixes Emissions, 2020. URL
https://www.zbe.barcelona/en/index.html.
ASPB. Agència de Salut Pública de Barcelona: Informe de qualitat de l’aire de Barcelona. Technical report,
2019.
ASPB. Agència de Salut Pública de Barcelona: Informe de qualitat de l’aire de Barcelona. Technical report,
2020. URL https://www.aspb.cat/documents/qualitat-aire-2020/.
R. Atkinson, B. Butland, H. Anderson, and R. Maynard. Long-term concentrations of nitrogen
dioxide and mortality: A meta-analysis of cohort studies. Epidemiology, 29:1, 05 2018. doi:
10.1097/EDE.0000000000000847.
J. M. Baldasano, M. T. Pay, O. Jorba, S. Gassó, and P. Jiménez-Guerrero. An annual assess-
ment of air quality with the CALIOPE modeling system over Spain. Science of the Total En-
vironment, 409(11):2163–2178, 2011. ISSN 00489697. doi: 10.1016/j.scitotenv.2011.01.041. URL
http://dx.doi.org/10.1016/j.scitotenv.2011.01.041.
J. Barceló. Fundamentals of Traffic Simulation. International Series in Operations Research & Management
Science. Springer New York, 2011. ISBN 9781441961426.
R. Beelen, G. Hoek, P. A. van den Brandt, R. A. Goldbohm, P. Fischer, L. J. Schouten, B. Armstrong, and
B. Brunekreef. Long-term exposure to traffic-related air pollution and lung cancer risk. Epidemiology, 19
(5):702–710, 2008. ISSN 1044-3983. doi: 10.1097/EDE.0b013e318181b3ca.
J. Benavides, M. Snyder, M. Guevara, A. Soret, C. Pérez Garćıa-Pando, F. Amato, X. Querol, and O. Jorba.
Caliope-urban v1.0: coupling r-line with a mesoscale air quality modelling system for urban air quality
forecasts over barcelona city (spain). Geoscientific Model Development, 12(7):2811–2835, 2019. doi:
10.5194/gmd-12-2811-2019. URL https://gmd.copernicus.org/articles/12/2811/2019/.
J. Benavides, M. Guevara, M. G. Snyder, D. Rodŕıguez-Rey, A. Soret, C. P. Garćıa-Pando, and O. Jorba.
22
On the impact of excess diesel NOX emissions upon NO2 pollution in a compact city. Environmental
Research Letters, 16(2), 2021a. ISSN 17489326. doi: 10.1088/1748-9326/abd5dd.
J. Benavides, M. Guevara, M. G. Snyder, D. Rodŕıguez-Rey, A. Soret, C. P. Garćıa-Pando, and O. Jorba.
Response 1, Reviewer 1, 2021b.
A. Bigazzi and M. Rouleau. Can traffic management strategies improve urban air quality? a review of the
evidence. Journal of transport and health, 7:111–124, 2017.
R. Borge, J. Lumbreras, J. Pérez, D. de la Paz, M. Vedrenne, J. M. de Andrés, and M. E.
Rodŕıguez. Emission inventories and modeling requirements for the development of air qual-
ity plans. application to madrid (spain). Science of The Total Environment, 466-467:809–
819, 2014. ISSN 0048-9697. doi: https://doi.org/10.1016/j.scitotenv.2013.07.093. URL
https://www.sciencedirect.com/science/article/pii/S0048969713008759.
R. Borge, J. L. Santiago, D. de la Paz, F. Mart́ın, J. Domingo, C. Valdés, B. Sánchez,
E. Rivas, M. T. Rozas, S. Lázaro, J. Pérez, and Álvaro Fernández. Application of a
short term air quality action plan in madrid (spain) under a high-pollution episode - part
ii: Assessment from multi-scale modelling. Science of The Total Environment, 635:1574 –
1584, 2018. ISSN 0048-9697. doi: https://doi.org/10.1016/j.scitotenv.2018.04.323. URL
http://www.sciencedirect.com/science/article/pii/S0048969718315018.
M. Brauer, G. Hoek, P. Van Vliet, K. Meliefste, P. H. Fischer, A. Wijga, L. P. Koopman, H. J. Neijens,
J. Gerritsen, M. Kerkhof, J. Heinrich, T. Bellander, and B. Brunekreef. Air pollution from traffic and the
development of respiratory infections and asthmatic and allergic symptoms in children. American journal
of respiratory and critical care medicine, 166(8):1092–1098, 2002. doi: 10.1164/rccm.200108-007OC.
D. Byun and K. Schere. Review of the governing equations, computational algorithms and other components
of the models-3 community multiscale air quality (cmaq) modeling system. Applied Mechanics Review,
59:51–78, 11 2005.
E. Croci. Urban road pricing: A comparative study on the experiences of
london, stockholm and milan. Transportation Research Procedia, 14:253–262,
2016. ISSN 2352-1465. doi: https://doi.org/10.1016/j.trpro.2016.05.062. URL
https://www.sciencedirect.com/science/article/pii/S235214651630062X. Transport Research
Arena TRA2016.
B. R. Denby, M. Gauss, P. Wind, Q. Mu, E. Grøtting Wærsted, H. Fagerli, A. Valdebenito, and H. Klein.
Description of the uemep v5 downscaling approach for the emep msc-w chemistry transport model.
Geoscientific Model Development, 13(12):6303–6323, 2020. doi: 10.5194/gmd-13-6303-2020. URL
https://gmd.copernicus.org/articles/13/6303/2020/.
DGT. DGT Tablas estadisticas, 2020. URL https://www.dgt.es/es/seguridad-vial/
estadisticas-e-indicadores/matriculaciones-definitivas/tablas-estadisticas/.
J. Duyzer, D. van den Hout, P. Zandveld, and S. van Ratingen. Representative-
ness of air quality monitoring networks. Atmospheric Environment, 104:88–101,
2015. ISSN 1352-2310. doi: https://doi.org/10.1016/j.atmosenv.2014.12.067. URL
https://www.sciencedirect.com/science/article/pii/S1352231014010152.
EEA. Air quality in Europe — 2019 report — EEA Report No 10/2019. Number 5. 2019. ISBN 978-92-
9480-088-6. doi: 10.2800/822355.
EMEP/EEA. Air pollutant emission inventory guidebook 2016 – Last Update June 2017. Technical report,
2017.
EMISIA. COPERT V, 2016. URL https://copert.emisia.com/.
European Automobile Manufacturers Association. Fuel types of new passenger cars, 2021. URL
https://www.acea.be/statistics/tag/category/share-of-diesel-in-new-passenger-cars.
R. Garćıa, J. Camaño, L. Pérez, and R. Barras. A health impact assessment of traffic restrictions during
madrid no 2 episode. IOP Conference Series: Earth and Environmental Science, 182:012003, 09 2018.
doi: 10.1088/1755-1315/182/1/012003.
A. Guenther, T. Karl, P. Harley, C. Wiedinmyer, P. I. Palmer, and C. Geron. Estimates of global
terrestrial isoprene emissions using megan (model of emissions of gases and aerosols from nature).
23
Atmospheric Chemistry and Physics, 6(11):3181–3210, 2006. doi: 10.5194/acp-6-3181-2006. URL
https://acp.copernicus.org/articles/6/3181/2006/.
M. Guevara, M. T. Pay, F. Mart́ınez, A. Soret, H. Denier van der Gon, and J. M.
Baldasano. Inter-comparison between hermesv2.0 and tno-macc-ii emission data us-
ing the caliope air quality system (spain). Atmospheric Environment, 98:134–145,
2014. ISSN 1352-2310. doi: https://doi.org/10.1016/j.atmosenv.2014.08.067. URL
https://www.sciencedirect.com/science/article/pii/S1352231014006694.
M. Guevara, C. Tena, M. Porquet, O. Jorba, and C. Pérez Garćıa-Pando. HERMESv3, a stand-alone
multiscale atmospheric emission modelling framework - Part 1: global and regional module. Geoscientific
Model Development Discussions, 12:1885–1970, 2019. doi: 10.5194/gmd-2018-324.
M. Guevara, C. Tena, M. Porquet, O. Jorba, and C. Pérez Garćıa-Pando. HERMESv3, a stand-alone multi-
scale atmospheric emission modelling framework – Part 2: The bottom–up module. Geoscientific Model
Development, 13:873–903, 2020. ISSN 1991-962X. doi: https://doi.org/10.5194/gmd-13-873-2020.
G. Hoek, P. Fischer, P. Van Den Brandit, S. Goldbohm, and B. Brunekreef. Estimation of long-term
average exposure to outdoor air pollution for a cohort study on mortality. Journal of Exposure Science
& Environmental Epidemiology, 11(6):459–469, 2001. ISSN 1559-064X. doi: 10.1038/sj.jea.7500189.
C. Holman, R. Harrison, and X. Querol. Review of the efficacy of low emission zones
to improve urban air quality in european cities. Atmospheric Environment, 111:161–
169, 2015. ISSN 1352-2310. doi: https://doi.org/10.1016/j.atmosenv.2015.04.009. URL
https://www.sciencedirect.com/science/article/pii/S1352231015300145.
C. Hood, I. MacKenzie, J. Stocker, K. Johnson, D. Carruthers, M. Vieno, and R. Doherty.
Air quality simulations for london using a coupled regional-to-local modelling system. Atmo-
spheric Chemistry and Physics, 18(15):11221–11245, 2018. doi: 10.5194/acp-18-11221-2018. URL
https://acp.copernicus.org/articles/18/11221/2018/.
A. Inness, M. Ades, A. Agusti-Panareda, J. Barre, A. Benedictow, A.-M. Blechschmidt, J. J. Dominguez,
R. Engelen, H. Eskes, J. Flemming, V. Huijnen, L. Jones, Z. Kipling, S. Massart, M. Parrington,
V. Peuch, M. Razinger, S. Remy, M. Schulz, and M. Suttie. The cams reanalysis of atmospheric composi-
tion. Atmospheric Chemistry and Physics, 19(6):3515–3556, 2019. doi: 10.5194/acp-19-3515-2019. URL
https://acp.copernicus.org/articles/19/3515/2019/.
Institut Cerdà. Assistència tècnica per al desplegament del Pla director de mo-
bilitat: Anàlisis del parc de vehicles, objectius i pla estratègic per al foment
de l’adquisició de vehicles de baixes emissions. Technical report, 2019. URL
https://doc.atm.cat/ca/ dir pdm estudis/Informe ParcVehiclesBaixesEmissions 31102019.pdf.
ISGlobal. ISGLobal Ranking of cities, 2021. URL https://isglobalranking.org/.
G. Janssens-Maenhout, M. Crippa, D. Guizzardi, F. Dentener, M. Muntean, G. Pouliot, T. Keating,
Q. Zhang, J. Kurokawa, R. Wankmuller, H. Denier van der Gon, J. J. P. Kuenen, Z. Klimont,
G. Frost, S. Darras, B. Koffi, and M. Li. Htap v2.2: a mosaic of regional and global emis-
sion grid maps for 2008 and 2010 to study hemispheric transport of air pollution. Atmo-
spheric Chemistry and Physics, 15(19):11411–11432, 2015. doi: 10.5194/acp-15-11411-2015. URL
https://acp.copernicus.org/articles/15/11411/2015/.
S. S. Jensen, R. Berkowicz, H. Sten Hansen, and O. Hertel. A danish decision-support gis tool for man-
agement of urban air quality and human exposures. Transportation Research Part D: Transport and
Environment, 6(4):229–241, 2001. ISSN 1361-9209. doi: https://doi.org/10.1016/S1361-9209(00)00026-2.
URL https://www.sciencedirect.com/science/article/pii/S1361920900000262.
S. S. Jensen, M. Ketzel, T. Becker, J. Christensen, J. Brandt, M. Plejdrup, M. Winther, O.-
K. Nielsen, O. Hertel, and T. Ellermann. High resolution multi-scale air quality modelling
for all streets in denmark. Transportation Research Part D: Transport and Environment, 52:
322 – 339, 2017. ISSN 1361-9209. doi: https://doi.org/10.1016/j.trd.2017.02.019. URL
http://www.sciencedirect.com/science/article/pii/S1361920916304072.
S. Khomenko, M. Cirach, E. Pereira-Barboza, N. Mueller, J. Barrera-Gómez, D. Rojas-Rueda, K. de Hoogh,
G. Hoek, and M. Nieuwenhuijsen. Premature mortality due to air pollution in European cities: a
24
health impact assessment. The Lancet Planetary Health, 3(5):e121 – e134, 2021. ISSN 2542-5196. doi:
10.1016/S2542-5196(20)30272-2.
J. J. P. Kuenen, A. J. H. Visschedijk, M. Jozwicka, and H. A. C. Denier van der Gon. TNO-MACC II
emission inventory; a multi-year (2003-2009) consistent high-resolution European emission inventory for
air quality modelling. Atmospheric Chemistry and Physics, 14(20):10963–10976, 2014. doi: 10.5194/acp-
14-10963-2014. URL https://acp.copernicus.org/articles/14/10963/2014/.
Kunz. KTT Umweltberatung und Software Street 5.2., 2005.
L. Montero, P. Linares, J. Salmerón, G. Recio, E. Lorente, and J. José Vázquez. Barcelona Virtual Mobility
Lab The multimodal transport simulation testbed for emerging mobility concepts evaluation. pages 2–5,
2018.
N. Mueller, D. Rojas-Rueda, H. Khreis, M. Cirach, D. Andrés, J. Ballester, X. Bartoll, C. Daher, A. Deluca,
C. Echave, C. Milà, S. Márquez, J. Palou, K. Pérez, C. Tonne, M. Stevenson, S. Rueda, and M. Nieuwen-
huijsen. Changing the urban design of cities for health: The superblock model. Environment Inter-
national, 134:105132, 2020. ISSN 0160-4120. doi: https://doi.org/10.1016/j.envint.2019.105132. URL
https://www.sciencedirect.com/science/article/pii/S0160412019315223.
NCEP. NOAA/NCEP Global Forecast System (GFS) Atmospheric Model, 2011.
M. Pay, F. Mart́ınez, M. Guevara, and J. Baldasano. Air quality forecasts on a kilometer-scale grid over
complex spanish terrains. Geoscientific Model Development, 7:1979–1999, 09 2014a. doi: 10.5194/gmd-
7-1979-2014.
M. T. Pay, F. Mart́ınez, M. Guevara, and J. M. Baldasano. Air quality forecasts on a kilometer-scale grid over
complex spanish terrains. Geoscientific Model Development, 7(5):1979–1999, 2014b. doi: 10.5194/gmd-
7-1979-2014. URL https://gmd.copernicus.org/articles/7/1979/2014/.
E. Pisoni, P. Christidis, P. Thunis, and M. Trombetti. Evaluating the impact of “sustainable ur-
ban mobility plans” on urban background air quality. Journal of Environmental Management,
231:249–255, 2019. ISSN 0301-4797. doi: https://doi.org/10.1016/j.jenvman.2018.10.039. URL
https://www.sciencedirect.com/science/article/pii/S0301479718311733.
Port de Barcelona. RePort, 2017. URL http://www.portdebarcelona.cat/en/web/el-port/report.
PTV Group. PTV VISUM 18 Manual, 2019. URL http://vision-traffic.ptvgroup.com/en-uk/
products/ptv-vissim/use-cases/emissions-modelling/.
D. Rodriguez-Rey, M. Guevara, M. P. Linares, J. Casanovas, J. Salmerón, A. Soret, O. Jorba,
C. Tena, and C. Pérez Garćıa-Pando. A coupled macroscopic traffic and pollutant emis-
sion modelling system for barcelona. Transportation Research Part D: Transport and Environ-
ment, 92:102725, 2021. ISSN 1361-9209. doi: https://doi.org/10.1016/j.trd.2021.102725. URL
http://www.sciencedirect.com/science/article/pii/S1361920921000274.
S. Rueda. Superblocks for the Design of New Cities and Renovation of Existing Ones: Barcelona’s
Case, pages 135–153. Springer International Publishing, Cham, 2019. ISBN 978-3-319-74983-9. doi:
10.1007/978-3-319-74983-9 8. URL https://doi.org/10.1007/978-3-319-74983-9 8.
W. C. Skamarock and J. B. Klemp. A time-split nonhydrostatic atmospheric model for
weather research and forecasting applications. Journal of Computational Physics, 227(7):
3465 – 3485, 2008. ISSN 0021-9991. doi: https://doi.org/10.1016/j.jcp.2007.01.037. URL
http://www.sciencedirect.com/science/article/pii/S0021999107000459. Predicting weather, cli-
mate and extreme events.
M. G. Snyder, A. Venkatram, D. K. Heist, S. G. Perry, W. B. Petersen, and V. Isakov.
Rline: A line source dispersion model for near-surface releases. Atmospheric Environment, 77:
748 – 756, 2013. ISSN 1352-2310. doi: https://doi.org/10.1016/j.atmosenv.2013.05.074. URL
http://www.sciencedirect.com/science/article/pii/S1352231013004470.
A. Soret, M. Guevara, and J. Baldasano. The potential impacts of electric vehicles on air
quality in the urban areas of barcelona and madrid (spain). Atmospheric Environment, 99:
51–63, 2014. ISSN 1352-2310. doi: https://doi.org/10.1016/j.atmosenv.2014.09.048. URL
https://www.sciencedirect.com/science/article/pii/S1352231014007419.
A. Valencia, A. Venkatram, D. Heist, D. Carruthers, and S. Arunachalam. Development
25
and evaluation of the r-line model algorithms to account for chemical transformation in the
near-road environment. Transportation Research Part D: Transport and Environment, 59:
464 – 477, 2018. ISSN 1361-9209. doi: https://doi.org/10.1016/j.trd.2018.01.028. URL
http://www.sciencedirect.com/science/article/pii/S1361920916308331.
26