sustainability
Article
Assessment of the Air Pollution Level in the City of
Rome (Italy)
Gabriele Battista *, Tiziano Pagliaroli, Luca Mauri, Carmine Basilicata
and Roberto De Lieto Vollaro
Department of Engineering, University of Roma TRE, via della Vasca Navale 79, Rome 00146, Italy;
[email protected] (T.P.); [email protected] (L.M.); [email protected] (C.B.);
[email protected] (R.D.L.V.)
* Correspondence: [email protected]; Tel.: +39-06-5733-3289
Academic Editor: Pietro Buzzini
Received: 20 May 2016; Accepted: 19 August 2016; Published: 23 August 2016
Abstract: Exposure to pollutants is usually higher in cities than in the countryside. Generally,
in the urban areas pollution sources as traffic, power generator and domestic heating system are
more intense and spatially distributed. The pollutants can be classified as a function of long-term
toxicological effects due to an exposure and inhalation. In the present work, several kinds of
pollutants concentration generated in Rome during 2015 have been analyzed applying different
advanced post-processing technique. In particular, statistic and cross-statistic have been computed
in time and phase space domain. As main result, it is observed, as expected, that all the pollutant
concentrations increase during the winter season into a couple of time ranges despite of [O3 ] that has
high values in summer. It can be clearly concluded that Rome has a strongly unsteady behaviour in
terms of a family of pollutant concentration, which fluctuate significantly. It is worth noticing that
there is a strong linear dependence between [C6 H6 ] and [NO] and a more complex interdependence
of [O3 ] and [C6 H6 ]. Qualitatively is provided that, to a reduction of [C6 H6 ] under a certain threshold
level corresponds an increase of [O3 ].
Keywords: air quality; emissions; pollutant; statistic analysis; domestic heating; urban traffic;
human exposure
1. Introduction
Outdoor air pollution has myriad sources, both natural and anthropogenic. It is a mixture
of mixtures, and the mix of contaminants in outdoor air varies widely in space and time,
reflecting variation in its sources, weather, atmospheric transformations and other factors. In any
particular place, the pollution in outdoor air comes not only from local sources, but also from sources
that affect air quality regionally and even globally [1]. The main reason for the air quality problems is
urban population growth, because people are constantly moving from rural to urban areas [2]. The air
pollutants result higher in urban area because of a combination of many elements such as industrial
activities, energy production plants and domestic heating [3]. They are noticeable by the toxicological
effects resulting from long-term exposure via inhalation [4].
The major pollutants produced are related to human activities, especially those produced by
combustion and industrial processes. The International Agency for Research on Cancer (IARC)
focused the attention on the human exposures to PM (Particulate Matter). The risk of being exposed to
a mixture of pollutants depends on particulate matter smaller than 2.5 µm, which is a common useful
indicator [4]. Therefore, the PM2.5 can be taken as indicator of population exposure to outdoor air
pollution. In 2010, 3.2 million of people died because of cardiovascular disease, caused by the exposure
Sustainability 2016, 8, 838; doi:10.3390/su8090838
www.mdpi.com/journal/sustainability
Sustainability 2016, 8, 838
2 of 15
to ambient fine particles (PM2.5 ) and 22,300 people died because of lung cancer. China and East Asia
show the largest number of people who lost their life [5,6].
With reference to NO2 , SO2 and (PMs) there is general agreement in the scientific literature
that they are the main agents responsible for the damage encountered on monuments and historical
buildings in urban areas [7]. Atmospheric composition is of unquestionable importance in the study of
the damage produced on building materials of artistic interest, since it directly influences the species
characteristics and entity of the degradation mechanism occurring on the cultural heritage.
The urban areas modified the environmental features that contributed to the increase of pollution.
As a matter of fact, the large concentration of the built environment, road pavement and the high
building materials capacitance changed the local micrometeorological conditions. Air temperature,
humidity and wind velocity and direction are altered in the urban environment compared to rural
areas. Furthermore, road traffic, domestic heating, industrial activities and lack buildings energy
performance involves high discomfort for users [8–20]. Besides the increase of pollutions, urbanization
has led to an increase of the urban heat island intensity (spatially-averaged surface or air-temperature
difference between an urban and surrounding rural area(s) [21]). Several studies are focused to reduce
the urban heat island effect with different mitigation techniques [22–28].
The people exposed to air pollutants are even more evident considering the weak ventilation,
because of the presence of high buildings with a consequent reduction of the dispersion of air masses.
For this reason, the contaminants formed below the building height remain at the pedestrian level
and increase the health damages especially during thermal inversion episodes. Several studies
were conducted to analyse the correlation between the street canyon features and the pollution
dispersion [29]. If the ratio between the average height of the buildings (H) and the width of
the canyon (W) is high enough to establish skimming flow conditions (at least higher than 0.65),
the retention of pollutants within the urban canopy layer will be amplified [30]. The major street
canyons in the cities have high value of the ratio H/W with a consequent established helical vortex
with an axis parallel to the canyon direction. In this case, the pollutants that go out of the canyon are
reduced [31].
The identification of analysis tools and methods, pollutant concentrations measurement,
comparison with the threshold values prescribed by law, are the activities foreseen by the legislations
in order to monitor the air quality and predict rehabilitation through the definition of plans
of interventions.
As a first step, in order to plain a control strategy of the pollution concentration in medium and
high scale cities, proper measurement and data processing are required to highlight the achievement
of dangerous concentration levels of pollutants and formulate a prediction model.
Actually, the main active control strategy is based on the introduction of some limits of the
urban traffic (e.g., number of the vehicles and vehicle categories that are authorized to transit).
Other interventions include the increase of efficiency of the heating systems in buildings such as the
replacement of traditional boilers with condensation ones integrated with more performing regulation
systems based on energy load tracking.
In the present work, several kinds of pollutants concentration such as [CO], [SO2 ], [NOx],
[NO], [NO2 ], [C6 H6 ], [PM10 ], [PM2.5 ] and [O3 ] generated in Rome during 2015 have been analysed.
These pollutants were taken from the Directive 2008/50/EC, the main legislation about ambient air
quality. In these analyses we applied different advanced post-processing techniques. Statistic and
cross-statistic have been computed in time and Fourier domain. In particular, probability distribution,
Kurtosis, Skewness, Poincaré sections and cross-correlation of the different pollutants were analysed
in order to assess the air pollution level in the city of Rome and the correlation of anthropogenic
sources with the pollutant emission. The extreme value theory was applied to the experimental
data. Especially using the generalized extreme value (GEV) distribution, several fittings of the
experiment probability density functions were calculated. GEV distribution introduced by Fisher and
Tippett [32] is commonly applied in environmental science to model a wide variety of natural extremes,
Sustainability 2016, 8, 838
3 of 15
including floods, rainfall, wind speeds, wave height, snow depths, earthquake, and other maxima.
For the present research activity, probability density functions fitting of the pollutant concentration
were calculated. The GEV distribution turned out to be very attractive mathematical tool since its
inverse has an analytical form, and its parameters are can be easily estimated [33–38]. This last feature
allows us to compute the return period: the likelihood of an event to occur. This post-processing
strategy is common applied to pollution database to develop model. Shen et al. developed a statistical
model using extreme value theory to estimate changes in ozone episodes [39].
2. Materials and Methods
2.1. Characteristics of the Study Area
Rome is the capital of Italy, and the largest city situated in the west-central part of the country.
It is also one of the most populous cities in the European Union with 2.9 million residents in 1285 km2 .
Considering the metropolitan area of Rome, the population reaches up to 4.3 million residents in
5352 km2 . With a history of more than two and half thousand years, Rome is called the Eternal City
because of the number of open-air museums. It is a mixture of a modern city and a plethora of
monuments, piazzas, villas, museums, churches, Egyptian obelisks, the Forum and Vatican City.
Its climate is typically Mediterranean: winters are cool and humid and summers are hot and dry. In the
coldest month (January) the temperature can reach about 0 ◦ C, and in the warmest months (July and
August) the temperature can reach 36 ◦ C.
The problem has two faces: there are a lot of pollutant activities and the public transport is not
efficient enough to reduce the number of cars. Pedestrian and public transport are only 20% of the total
mobility, while 60% of journeys are made by private means of transport; in the historical centre this
modal share changes into 34% of pedestrians, 29% of public transport and 37% of private transport [40].
In order to partially solve the problem of pollution, the Government has developed a lot of policies to
improve the net of the means of transport [41].
Studying the air pollutant in the city of Rome is important because there is a high density of
population and for the effects of pollutant on the variety of open-air museum.
2.2. Pollutant Legislation
The European Union policy on air quality has the goal of implementing appropriate instruments
aimed at improving air quality. The control of emissions from mobile sources, improving fuel quality
and promoting and integrating environmental protection requirements into the transport and energy
sector are the main goals of this policy [42].
The Directive 2008/50/EC of the European Parliament [43] is the main legislation about
ambient air quality. It fixes health based standards and objectives for a number of pollutants in
air. These standards are applied in different time spans because the observed health impacts associated
with the various pollutants occur over different exposure times. Table 1 shows the standard adopted
in the Directive 2008/50/EC.
Table 1. Directive 2008/50/EC standards of main pollutant.
Pollutant
PM10
PM10
PM2.5
NO2
NO2
SO2
SO2
O3
CO
C6 H6
Concentration
µg/m3
50
40 µg/m3
25 µg/m3
200 µg/m3
40 µg/m3
350 µg/m3
125 µg/m3
120 µg/m3
10 mg/m3
5 µg/m3
Averaging Period
Permitted Excess Each Year
24 h
1 year
1 year
1h
1 year
1h
24 h
Maximum daily 8 h mean
Maximum daily 8 h mean
1 year
35
18
24
3
25 days averaged over 3 years
-
Sustainability 2016, 8, 838
4 of 15
2.3. Post Processing Techniques
In this paper, different statistics and cross-statistics have been computed in time and Fourier
domain. These correspond to time-frequency analysis that study a signal in both the time and the
frequency domain simultaneously. In particular, the different techniques adopted are:
•
•
•
•
•
•
•
Probability distribution describes the possible values that a random variable can take within
a given range.
Kurtosis is a measure of whether the data have a flattening or elongation from the normal
distribution. High kurtosis indicates a flattering distribution, while low values indicate
an elongation distribution.
Skewness is a measure of the asymmetry of the distribution. A data set is symmetric if it looks
the same to the left and right of the center point.
Poincaré sections are a way to represent a dynamical system. The surface of section presents
a trajectory in n-dimensional phase space in an (n-1)-dimensional space. By picking one phase
element constant and plotting the values of the other elements each time the selected element has
the desired value, an intersection surface is obtained. The phase space is a surface that describes
all the possible states of a system.
Cross-correlation is a measure of similarity of two data series as a function of the lag of one relative
to the other.
Coefficient of variation normalizes the standard deviation with the mean of a data. This index
gives information about the variability of a data set.
Generalized Extreme Value distribution is often applied to analyse a large set of data
characterized by small or large value. In this approach three simpler distributions into a single
form are combined, allowing a continuous range of possible shapes.
2.4. Monitoring Station Network
ARPA Lazio [44], the regional environmental agency, operates several air quality monitoring sites
in the Lazio Region, including Rome. The Rome monitoring network consists of 10 monitoring stations
of [CO], [SO2 ], [NOx], [NO], [NO2 ], [C6 H6 ], [PM10 ], [PM2.5 ] and [O3 ] that are shown in Figure 1.
The monitoring network acquires concentration data every hour and every stations is set to monitor
different type of concentrations. All the stations taken into account are placed inside high density
urban areas that are characterized by traffic and domestic heating system pollutant sources.
The monitoring stations are constituted by fixed structures in which the entire measurement
instrument, the acquisition systems and local storage are placed inside. The procedures and the control
of the quality of the measures are assured by ARPA Lazio [45]. A zero value was considered when
an instruments errors were found.
All the measurement instruments are made by Project Automation. Table 2 shows the sensors
used for each concentration.
Table 2. Sensors used for each concentration measurement.
Pollutant
PMs
NOx
MP101MC
SWAMDC FAI
Sensors
SWAM5a FAI
SWAM DC FAI
TE SHARP 5030
SO2
O3
TE 43i
M200 A-API
CO
C6 H6
TE 48i
AIR Toxic
M300E API
CP 7001
M400E API
M100E API
Sustainability 2016, 8, 838
Sustainability 2016, 8, 838
5 of 15
5 of 15
Figure
1. Google
Earth
map
Rome.Red
Red circles
circles are
stations
taken
intointo
account.
Figure
1. Google
Earth
map
ofofRome.
arethe
themonitoring
monitoring
stations
taken
account.
3. Results and Discussion
3. Results and Discussion
In the present study, the data recorded in 2015 were considered. It is useful to consider the
In
the present
study,
the data recorded
2015 were
considered.
It is useful
to consider
the mean
mean
values of
contaminants
of each in
station
in order
to represent
the average
pollutant
values
of contaminants
of each station in order to represent the average pollutant concentrations
concentrations
of Rome.
of Rome. Before calculating the mean values concentrations of Rome, Figures 2 and 3 show the mean,
standardcalculating
deviation, Kurtosis
andvalues
Skewness
values of eachof
pollutant
the2 year
These
aremean,
Before
the mean
concentrations
Rome, during
Figures
and2015.
3 show
the
analysed
in
order
to
find
the
behaviour
of
each
pollutant
in
the
different
stations.
In
the
case
of are
standard deviation, Kurtosis and Skewness values of each pollutant during the year 2015. These
[PM10], there is only one station that indicates a different Kurtosis value than the others. Despite this
analysed in order to find the behaviour of each pollutant in the different stations. In the case of
consideration, in all the concentrations the Kurtosis and Skewness values of the different stations are
[PM10 ], there is only one station that indicates a different Kurtosis value than the others. Despite this
similar and it suggest that the mean of the network is possible.
consideration,
in all the
concentrations
Kurtosis
and Skewness
values of
theduring
different
From Figures
4–11
is shown how the
the mean
concentration
of all stations
work
the stations
year in are
similar
and
it
suggest
that
the
mean
of
the
network
is
possible.
each hour. It can been seen that the major pollutant concentrations happen in the winter seasons
From
4–11
shown
howthere
the mean
concentration
during
from 8Figures
to 13, and
fromis17
to 2, when
is a greater
flow of carsof
inall
thestations
city. Thework
highest
values the
are year
in each
hour. from
It can19been
seen
that
majorsystems
pollutant
concentrations
happen
in there
the winter
seasons
recorded
to 23,
when
thethe
heating
are turned
on. During
the night
are lower
values
due
to
the
turned
off
heating
systems
that
is
usually
in
Italy.
In
the
summer,
the
high
values
from 8 to 13, and from 17 to 2, when there is a greater flow of cars in the city. The highest values are
in these
hours
duewhen
to the the
urban
traffic. systems are turned on. During the night there are lower
recorded
from
19 are
to 23,
heating
It
is
worth
noticing
that
the
[CO],
[NO],
[NO
] and [C6in
H6Italy.
] haveInsimilar
trend during
the year
values due to the turned off heating systems
that
is 2usually
the summer,
the high
values in
suggesting the presence of pollutant source simultaneously.
these hours are due to the urban traffic.
The [O3] is recorded to be higher in the summer and during the hottest hours, when the solar
It is worth noticing that the [CO], [NO], [NO2 ] and [C6 H6 ] have similar trend during the year
radiation is highest. As a matter of fact, the presence of solar radiation allows the reaction of nitrogen
suggesting
presence
of pollutant source simultaneously.
dioxidethe
(NO
2) in the formation of ozone (O3).
The The
[O3 ][SO
is recorded
to be higher in the summer and during the hottest hours, when the solar
2] have values up to 8 μg/m3 compared with the limit of 125 μg/m3 reported in Table 1.
radiation
is
highest.
As
a
matter
fact,
presence
radiation allows
thepower
reaction
of and
nitrogen
That is foreseeable because theof
[SO
2] isthe
mainly
causedofbysolar
the combustion
of fuel at
plants
dioxide
(NO
formation
ozone
other
industrial
facilities.
As aof
matter
of(O
fact,
2 ) in the
3 ). in the center of Rome there is not this kind of system.
3 compared
3 reported
The 2concentration
and PM
10 has highwith
values
the of
winter
season
and reachinthe
The [SO
] have valuesof
upPM
to 2.58 µg/m
theinlimit
125 µg/m
Table 1.
3
values
in the last
of 2015,
with by
values
close to 50 μg/m
of PM
2.5 and values
That maximum
is foreseeable
because
the two
[SO2months
] is mainly
caused
the combustion
of fuel
at power
plants and
3 of PM10.
to 70 μg/m
otherclose
industrial
facilities.
As a matter of fact, in the center of Rome there is not this kind of system.
The concentration of PM2.5 and PM10 has high values in the winter season and reach the maximum
values in the last two months of 2015, with values close to 50 µg/m3 of PM2.5 and values close to
70 µg/m3 of PM10 .
Sustainability 2016, 8, 838
Sustainability 2016, 8, 838
Sustainability 2016, 8, 838
6 of 15
6 of 15
6 of 15
Figure
2. Mean
(blue)
and
standarddeviation
deviation (red) values
allall
pollutant
species
in the
Figure
2. Mean
(blue)
and
standard
valuescalculated
calculatedfor
pollutant
species
in the
Figure
2. Mean
(blue)
and
standard deviation (red)
(red) values
calculated
forfor
all pollutant
species
in the
different
stations.
Blue
line
is
the
mean
of
the
mean
values
of
the
different
stations
and
the
red
line
is
different
stations.
Blue
line
is
the
mean
of
the
mean
values
of
the
different
stations
and
the
red
different stations. Blue line is the mean of the mean values of the different stations and the red line is line
the mean of the standard deviation values of the different stations. (a) [CO]; (b) [NO]; (c) [NO2];
is the
valuesofofthe
thedifferent
differentstations.
stations.
[CO];
[NO];
(c) [NO
2 ];
2];
themean
meanofofthe
thestandard
standard deviation
deviation values
(a) (a)
[CO];
(b) (b)
[NO];
(c) [NO
(d) [C6H6]; (e) [O3]; (f) [SO2]; (g) [PM2.5]; (h) [PM10].
(d) [C
H
];
(e)
[O
];
(f)
[SO
];
(g)
[PM
];
(h)
[PM
].
2.5]; (h) [PM10].
(d)6 [C66H6]; (e) [O
3 3]; (f) [SO
2 2]; (g) [PM2.5
10
Figure 3. Kurtosis (blue) and Skewness (red) values calculated for all pollutant species in the
Figure
3. Kurtosis
(blue) and
Skewness (red)
values
calculated all
forpollutant
all pollutant
species
in the
Figure
3. Kurtosis
(blue)
Skewness
calculated
species
different
stations.
Blue and
line is
the mean(red)
of thevalues
Kurtosis
values of for
the different
stations
and in
thethe
reddifferent
line
different
stations.
Blue
line
is
the
mean
of
the
Kurtosis
values
of
the
different
stations
and
the
red
line the
stations.
is Skewness
the meanvalues
of theof
Kurtosis
values
of the(a)
different
stations
and
the
line
is the Blue
meanline
of the
the different
stations.
[CO]; (b)
[NO]; (c)
[NO
2]; red
(d) [C
6H6is
];
is the mean of the Skewness values of the different stations. (a) [CO]; (b) [NO]; (c) [NO2]; (d) [C6H6];
mean
Skewness
the[PM
different
stations. (a) [CO]; (b) [NO]; (c) [NO2 ]; (d) [C6 H6 ]; (e) [O3 ];
(e)of[Othe
3]; (f)
[SO2]; (g)values
[PM2.5];of(h)
10].
(e) [O3]; (f) [SO2]; (g) [PM2.5]; (h) [PM10].
(f) [SO2 ]; (g) [PM2.5 ]; (h) [PM10 ].
Sustainability 2016, 8, 838
Sustainability
2016,
838
Sustainability 2016,
2016, 8,
8, 838
838
Sustainability
Sustainability 2016,8,
8, 838
hour
hour
hour
hour
24
24
24
22
24
22
22
20
22
20
20
18
20
18
18
16
18
16
16
14
16
14
14
12
14
12
12
10
12
10
10
8
10
8
868
6
646
4
424
2
22
7 of 15
77 of
15
of 15
15
77of
of 15
30
30
30
30
60
60
60
60
90
90
90
90
120
120
120
120
150
150
150
150
180
180
day
180
180
day
day
day
210
210
210
210
240
240
240
240
270
270
270
270
300
300
300
300
330
330
330
330
360
360
360
360
12
12
12
12
10
10
10
10
8
8
88
6
6
66
4
4
44
2
2
22
0
0
00
Figure
µg/m333.. .
Figure
4.
Hourly
annual
concentration
variation
of
C
during
in
Figure 4.
4. Hourly
Hourly annual
annual concentration
concentration variation
variationof
ofC
C666H
H6666 during
during 2015
2015 in
in μg/m
μg/m
Figure
Figure 4.
4. Hourly
Hourly annual
annual concentration
concentration variation
variation of
of CC66H
H6 during 2015
2015 in μg/m
μg/m33..
hour
hour
hour
hour
24
24
24
24
22
22
22
20
22
20
20
18
20
18
18
18
16
16
16
14
16
14
14
12
14
12
12
10
12
10
10
10
8
8
868
6
646
4
424
2
22
30
30
30
30
60
60
60
60
90
90
90
90
120
120
120
120
150
150
150
150
180
180
day
180
180
day
day
day
210
210
210
210
240
240
240
240
270
270
270
270
300
300
300
300
330
330
330
330
360
360
360
360
3.5
3.5
3.5
3.5
3
3
33
2.5
2.5
2.5
2.5
2
2
22
1.5
1.5
1.5
1.5
1
1
11
0.5
0.5
0.5
0.5
0
0
00
3
Figure
5.
annual
variation
CO
2015
in
mg/m
Figure 5.
5. Hourly
Hourly annual
annual concentration
concentration variation
variation of
of CO
CO during
during 2015
2015 in
in mg/m
mg/m3333....
Figure
Figure 5.Hourly
Hourly annualconcentration
concentration variationof
of COduring
during 2015in
in mg/m
mg/m .
400
400
400
400
300
300
300
300
hour
hour
hour
hour
24
24
24
22
24
22
22
20
22
20
20
18
20
18
18
16
18
16
16
14
16
14
14
12
14
12
12
10
12
10
10
8
10
8
868
6
646
4
424
2
22
200
200
200
200
100
100
100
100
30
30
30
30
60
60
60
60
90
90
90
90
120
120
120
120
150
150
150
150
180
180
day
180
180
day
day
day
210
210
210
210
240
240
240
240
270
270
270
270
300
300
300
300
330
330
330
330
360
360
360
360
0
0
00
3
Figure
6.
annual
concentration
variation
of
NO
during
2015
in
μg/m
Figure 6.
6. Hourly
Hourly annual
annual concentration
concentration variation
variation of
of NO
NO during
during 2015
2015 in
in μg/m
μg/m333...
Figure
Figure6.6. Hourly
Hourly annual
annualconcentration
concentrationvariation
variationof
ofNO
NOduring
during2015
2015in
inµg/m
μg/m .
Figure
Hourly
hour
hour
hour
hour
24
24
24
24
22
22
22
22
20
20
20
20
18
18
18
18
16
16
16
16
14
14
14
14
12
12
12
12
10
10
10
10
8
8688
6466
4244
222
30
30
30
30
60
60
60
60
90
90
90
90
120
120
120
120
150
150
150
150
180
180
day
180
180
day
day
day
210
210
210
210
240
240
240
240
270
270
270
270
300
300
300
300
330
330
330
330
3
Figure
7.
Hourly
annual
concentration
variation
of
NO
during
2015
in
μg/m
Figure 7.
7. Hourly
Hourly annual
annual concentration
concentration variation
variation of
of NO
NO222 during
during 2015
2015 in
in μg/m
μg/m3333...
Figure
Figure7.7.Hourly
Hourlyannual
annualconcentration
concentrationvariation
variationof
ofNO
NO22 during
during 2015
2015in
inµg/m
μg/m ..
Figure
360
360
360
360
160
160
160
160
140
140
140
140
120
120
120
120
100
100
100
100
80
80
80
80
60
60
60
60
40
40
40
40
20
20
20
20
0
000
Sustainability 2016, 8, 838
Sustainability 2016, 8, 838
8 of 15
8 of 15
hour
hour
8 of 15
8 of160
15
hour
hour
Sustainability 2016, 8, 838
Sustainability
2016, 8, 838
24
22
24
20 22
24
18 20
22
16 18
20
14 16
18
12 14
16
10 12
14
12
8 10
8
6 10
4 68
2 46
24
30
60
2
30
30
60
60
90
90
90
120
120
120
150
150
150
180
day
180
day
180
day
210
210
210
240
240
240
270
270
270
300
300
300
330
330
330
160140
160
140120
140
120100
120
10080
100
80 60
80
60
40
60
40
40 20
20
360 20
360
360
24
24
22 22
24
20 20
22
18 18
20
16 16
18
14 14
16
12 12
14
12
10 10
8 10
8
6 68
4 46
2 24
Figure 8. Hourly annual concentration variation of O3 during 2015 in μg/m3.
8
8
8
6 6
6
hour
hour
hour
hour
33. 3
Figure
Hourlyannual
annualconcentration
concentration variation of
O
during
2015
ininμg/m
Figure
8. 8.
Hourly
ofO
O3333during
during
2015
µg/m
.
3.
Figure
8. Hourly
annual concentrationvariation
variation of
2015
in μg/m
2
4 4
4
2 2
2
30 30
30
6060
60
9090
90
120
120
120
150
150
150
180
180
day
180
day
day
210
210
240
240
240
270
270
270
300
300
300
330
330
330
Figure
9.
Hourly
annual
concentrationvariation
variation of
2015
ininμg/m
. 33. 3 .
Figure
Hourly
annual
concentration
variation
of SO
SO
during
2015
Figure
9. 9.
Hourly
annual
concentration
SO22222during
during
2015
inμg/m
µg/m
3.
Figure
9. Hourly
annual
concentration variation of SO
during
2015
in μg/m
3
Figure 10. Hourly annual concentration variation of PM2.5 during 2015 in μg/m3.
Figure
10. Hourly
annual concentrationvariation
variation of PM2.5 during
2015
in μg/m
. 3 3.
Figure
10.10.
Hourly
during
2015
in µg/m
Figure
Hourlyannual
annualconcentration
concentration variation of
2.5
of PM2.5
2.5 during 2015 in μg/m 3.
3
Figure 11. Hourly annual concentration variation of PM10 during 2015 in μg/m3.
Figure 11. Hourly annual concentration variation of PM10 during 2015 in μg/m3.
3
Figure
Hourlyannual
annualconcentration
concentrationvariation
variation of
of PM
10 during 2015 in μg/m 3. 3
Figure
11.11.
Hourly
PM10
10 during 2015 in µg/m .
0 0
360
360 0
360
Sustainability 2016, 8, 838
Sustainability 2016, 8, 838
9 of 15
9 of 15
The
main
pollutant
species
terms
probability
density
function
(pdf)
have
been
analysed
The
main
pollutant
species
in in
terms
of of
probability
density
function
(pdf
) have
been
analysed
and
and
represented
in
normalized
form
in
order
to
have
zero
mean
and
unitary
standard
deviation
represented in normalized form in order to have zero mean and unitary standard deviation defined
as defined
follows:as follows:
x − hxi
xnorm == − 〈 〉
(1) (1)
σ
σx
where
as as
h·i〈∙〉
is denoted
the
ensemble
deviation.
where
is denoted
the
ensembleaverage
averageand
andas
as σσisisindicated
indicatedthe
the standard
standard deviation.
It It
is is
shown
in
Figure
12
that
[CO],
[NO],
[SO
]
and
[C
H
]
have
a
narrow
distribution
around
2
6
6
shown in Figure 12 that [CO], [NO], [SO2] and [C6H6] have a narrow distribution
around
the
themost
mostfrequent
frequentvalue
valueand
andsignificant
significant
positive
tail.
The
latter
aspect
can
be
ascribed
to
intermittent
positive tail. The latter aspect can be ascribed to intermittent
energetic
events
embedded
in in
concentration
time
histories.
To To
better
clarify
thisthis
aspect
all pdf
are are
fitted
energetic
events
embedded
concentration
time
histories.
better
clarify
aspect
all pdf
using
the
Generalized
Extreme
Value
(GEV)
function.
By
means
of
the
GEV
function,
the
return
level
fitted using the Generalized Extreme Value (GEV) function. By means of the GEV function, the
and
period
were
in calculated
order to evaluate
occurrence
time of these
events
as reported
return
level
andcalculated
period were
in order the
to evaluate
the occurrence
time
of these
events as in
reported
in Figure
13. With
reference
the return
period
of [NO],
every
two days
a high
energetic
Figure
13. With
reference
to the
return to
period
of [NO],
every
two days
a high
energetic
peak
in [NO]
peak in [NO]can
concentration
On the
other
hand,
in
[NO
2
]
time
history
intermittent
concentration
be detected.can
Onbe
thedetected.
other hand,
in [NO
]
time
history
intermittent
events
are
rarely
2
events
are
rarely
observed.
This
aspect
is
a
footprint
of
chaotic
behavior
of
the
Rome
in
terms
of
observed. This aspect is a footprint of chaotic behavior of the Rome in terms of pollutant generation
pollutant
generation
could during
be takenthe
into
account during
thethis
modelling
and
could be
taken intoand
account
modelling
stage of
process.stage of this process.
Instead
[NO
2],[PM
[PM
2.5],
],
[PM
10]] and
[O
3] ]have
a
distribution
with
a more
flatflat
trend.
In In
particular,
Instead
[NO
],
[PM
and
[O
have
a
distribution
with
a more
trend.
particular,
2
2.5
10
3
[NO
2] pdf is well fitted by the Gaussian pdf except for a small region over 3σ where a departure of the
[NO2 ] pdf is well fitted by the Gaussian pdf except for a small region over 3σ where a departure of
In In
thethe
latter
case,
wewe
can
consider,
in in
a first
approximation,
thepositive
positivetail
tailcan
canbebeclearly
clearlyobserved.
observed.
latter
case,
can
consider,
a first
approximation,
equally probable concentrations in a certain range.
equally probable concentrations in a certain range.
101
101
100
100
10
-1
10-1
10
-2
10-2
10-3
10-3
10-4
-2
0
2
4
6
8
10
10-4
-2
101
101
0
100
10-1
10-1
10-2
10-2
10
-3
10-3
10
-4
10
-2
0
2
4
6
8
10
10-4
-2
0
2
4
6
8
10
0
2
4
6
8
10
Figure12.
12. Probability
(red(red
triangles)
and Generalized
Extreme
Value fitting
Figure
Probabilitydensity
densityfunction
function
triangles)
and Generalized
Extreme
Value(blue
fitting
circles)
of pollutant
concentrations
calculated
for samples
acquired
in a year.
(blue
circles)
of pollutant
concentrations
calculated
for samples
acquired
in a year.
Sustainability 2016, 8, 838
10 of 15
Sustainability 2016, 8, 838
10 of 15
Analysing the probability distributions in comparing with the standards reported in
Table 1,
it could be
notice
that
[NO
],
[PM
],
[PM
],
[C
H
]
and
[O
]
exceeds
the
limit
of
legislation
2
2.5
10 in comparing
6 6
3 the standards reported in Table 1,during
Analysing the probability
distributions
with
it
3 , 23% of time the [PM ]
the year.
In
particular,
64%
of
time
the
[NO
]
was
over
the
limit
of
40
µg/m
2
could be notice that [NO2], [PM2.5], [PM10], [C6H6] and [O3] exceeds the limit of legislation during the 2.5
3 , and 21% of time [PM ] was over the limit
3 Regarding the
was over
limit of 2564%
µg/m
of of
40time
µg/m
10 limit of 40 μg/m3, 23%
year. the
In particular,
of time
the [NO2] was over the
the .[PM
2.5] was
3. Regarding
concentration
of [Cof6 H
and 3[O
time
in which
concentration
the limit
of legislation
over the limit
256 ]μg/m
, and
21%
of time
[PM10]the
was
over the limitisofover
40 μg/m
the is
3 ] the
3
concentration
[C6H6] and4%
[Oof
3] the
time
the concentration
is of
over
limit and
of legislation
is the
less than
previousofpollutant:
time
thein[Cwhich
over the limit
40the
µg/m
3% of time
6 H6 ] was
pollutant:
4% of3time
the [C6H6] was over the limit of 40 μg/m3 and 3% of time the
[O3 ] less
wasthan
overprevious
the limit
of 120 µg/m
.
3
[O
] was over
the limit
120 μg/m
In32015,
regarding
theoflimit
in the. short averaging period (from 1 to 24 h), [NO2 ], [SO2 ] and [CO]
In 2015, regarding
the limit
in the shortFor
averaging
(from
1 toµg/m
24 h),3[NO
], [SO
2] and [CO]
haven’t exceeded
the legislation
thresholds.
[PM10 ]period
the limit
of 50
for 2an
averaging
period
haven’t exceeded the legislation thresholds. For [PM10] the limit of 50 μg/m3 for an averaging period
of 24 h was exceeded 39 times that is more than 35 permitted exceedances each year by the legislation.
of 24 h was exceeded 39 times that is more than 35 permitted exceedances each year by the
These exceedances are concentrated in December when there was a combination of exogenous state
legislation. These exceedances are concentrated in December when there was a combination of
variable
(i.e., temperature,
humidity
and wind
velocity
and
traffic flow)
and low
exogenous
state variable (i.e.,
temperature,
humidity
and and
winddirection,
velocity and
direction,
and traffic
3 for an averaging period of 8 h was exceeded 7 times. For this
rainfall.
For
[O
]
the
limit
of
120
µg/m
3 rainfall. For [O3] the limit of 120 μg/m3 for an averaging period of 8 h was exceeded 7
flow) and low
concentration,
the concentration,
legislation permitted
25 exceedances
over 3 years.
times. For this
the legislation
permitted 25days
exceedances
days over 3 years.
Figure 13. Dimensionless Return level trend for all the pollutant concentrations.
Figure 13. Dimensionless Return level trend for all the pollutant concentrations.
As stated previously and as further confirmed by the values assumed by Kurtosis and
As stated (see
previously
asreference
further confirmed
by[CO],
the values
and
Skewness
Skewness
Table 3),and
with
to the set of
[NO], assumed
[SO2] and by
[C6Kurtosis
H6] a high
Kurtosis
value
is
evaluated,
indicating
that
these
pollutants
are
present
on
Rome
in
constant
quantities.
On
(see Table 3), with reference to the set of [CO], [NO], [SO2 ] and [C6 H6 ] a high Kurtosis value is evaluated,
the other
thepollutants
high Kurtosis
values suggests
distribution
around
theother
mode.hand
Whereas
indicating
thathand
these
are present
on Romea sharped
in constant
quantities.
On the
the high
the coefficient
of variation
of [NO]
have large value
indicating
thatWhereas
there are the
highcoefficient
energetic extreme
Kurtosis
values suggests
a sharped
distribution
around
the mode.
of variation
value
(seelarge
Figurevalue
12). indicating that there are high energetic extreme value (see Figure 12).
of [NO]
have
Table 3. Kurtosis and Skewness values calculated for all pollutant species.
Table 3. Kurtosis and Skewness values calculated for all pollutant species.
Specie
[CO]
[NO]
[SO2]
[C6H6]
[NO2]
[PM2.5]
[PM10]
[O3]
Kurtosis
14.3
Kurtosis
17.1
[CO]
14.3
9.6
[NO]
17.1
9.4
[SO2 ]
9.6
3.9
[C6 H6 ]
9.4
4.2
[NO2 ]
3.9
3.7
[PM2.5 ] 2.8 4.2
Specie
[PM10 ]
[O3 ]
3.7
2.8
Skewness
Coefficient of Variation
2.7 Coefficient of Variation
66%
Skewness
2.7
3.3
2.2
2.1
0.7
1.3
1.0
0.7
3.3
2.2
2.1
0.7
1.3
1.0
0.7
66%
155%
91%
72%
48%
60%
46%
80%
155%
91%
72%
48%
60%
46%
80%
Sustainability 2016, 8, 838
11 of 15
In the case of [NO2 ] the values assumed by the Kurtosis are close to those of a Gaussian
Sustainability 2016, 8, 838
11 of 15
Sustainability
8, 838
of 15
distribution
as2016,
confirmed
by the comparison in Figure 12, indicating the presence of a 11
species
of
pollutant In
that
significantly
around
an average
thefluctuate
case of [NO
2] the values
assumed
by the value.
Kurtosis are close to those of a Gaussian
In the case of [NO2] the values assumed by the Kurtosis are close to those of a Gaussian
Given
that the
most important
benzene source
is the12,urban
traffic,
concentration
of benzene
distribution
as confirmed
by the comparison
in Figure
indicating
thethe
presence
of a species
of
distribution as confirmed by the comparison in Figure 12, indicating the presence of a species of
that
anits
average
value. can be taken into account as reference.
may pollutant
be used as
anfluctuate
index ofsignificantly
city traffic around
flux and
time history
pollutant that fluctuate significantly around an average value.
Given that
the most important
benzeneby
source
is the
urban traffic,
the concentration
benzene
The production
of benzene
is characterized
small
oscillations
in Rome
during theof
window
Given that
the most important
benzene source
is the
urban traffic,
the concentration
oftime
benzene
may
be
used
as
an
index
of
city
traffic
flux
and
its
time
history
can
be
taken
into
account
as
reference.
investigated.
Furthermore,
[C
H
]
are
strongly
related
to
the
concentration
of
pollutants
species
6 traffic
6
may be used as an index of city
flux and its time history can be taken into account as reference.
The production of benzene is characterized by small oscillations in Rome during the time window
such The
as [CO],
[NO] of
and
[NO2 ],
confirmedby
bysmall
the cross-correlation
values
in Figure 14.
production
benzene
is as
characterized
oscillations in Rome
duringreported
the time window
investigated. Furthermore, [C6H6] are strongly related to the concentration of pollutants species such
investigated.isFurthermore,
[C6H
6] are
stronglyoxide
relatedand
to the
concentration
of pollutants
species
such and
This correlation
due to the fact
that
nitrogen
carbon
oxides are
combustion
products
as [CO], [NO] and [NO2], as confirmed by the cross-correlation values reported in Figure 14. This
as [CO],
[NO] and are
[NOdue
2], asto
confirmed
by the cross-correlation
values
reported
in Figure 14.
This
that their
fluctuations
the
oscillations
of
the
city
traffic
flux.
It
is
noteworthy
that
[O ] is
correlation is due to the fact that nitrogen oxide and carbon oxides are combustion products and that 3
correlation
is
due
to
the
fact
that
nitrogen
oxide
and
carbon
oxides
are
combustion
products
and
that
weakly
related
to theare
production
of benzeneofbut
is traffic
indeed
in It
phase
opposition
To further
their
fluctuations
due to the oscillations
the it
city
flux.
is noteworthy
thatwith
[O3] isit.weakly
their fluctuations
are due
to the the
oscillations
of section
the city traffic
flux.
It is noteworthy
that
[O3]] upon
is weakly
investigate
this
interesting
aspect,
Poincaré
of
[C
H
]
upon
[NO]
and
[C
H
[O3 ] are
6
related to the production of benzene but it is indeed in6 phase
opposition with 6it. 6To further
related to the production of benzene but it is indeed in phase opposition with it. To further
represented
in
Figure
15.
As
noted
above
there
is
a
strong
linear
dependence
between
[C
H
]
and
investigate this interesting aspect, the Poincaré section of [C6H6] upon [NO] and [C6H6] upon
6 [O
6 3] are [NO]
investigate this interesting aspect, the Poincaré section of [C6H6] upon [NO] and [C6H6] upon [O3] are
in Figure
15. As noted of
above
strong
linear
dependence
6H6] and
and arepresented
more complex
interdependence
[O ]there
and is
[Ca H
specifically
anbetween
inverse [C
proportionality
6 ]. More
represented in Figure 15. As noted above3 there is a6 strong
linear dependence between [C6H6] and
[NO]observed
and a more
complex interdependence
of [O3] and [C6H6]. More specifically an inverse
is clearly
as
formalized
in
the
following:
[NO] and a more complex interdependence of [O3] and [C6H6]. More specifically an inverse
proportionality is clearly observed as formalized in the following:
proportionality is clearly observed asformalized in thefollowing:
[O
d [C[C
d [O
3 ] 6 HH6 ]]
]
∼−− [O ] [C H ] ~
dt
dt
~−
(2)
(2)
Figure
14. Cross-correlation
betweenpollutant
pollutant concentrations;
concentrations; benzene
concentration
is taken
into into
Figure
14. Cross-correlation
between
benzene
concentration
is taken
Figure 14. Cross-correlation between pollutant concentrations; benzene concentration is taken into
account
as
a reference
signal;
the
timeaxis
axisisisnormalized
normalized respect
totoa a
reference
time:
t* = t*
86,400
s.
account
as
a
reference
signal;
the
time
respect
reference
time:
=
86,400
account as a reference signal; the time axis is normalized respect to a reference time: t* = 86,400 s. s.
Figure
15. Poincaré
sections[C[CH
6H6] over [NO] (a) and [C6H6] over [O3] (b).
Figure
15. Poincaré
sections
[NO](a)
(a)and
and[C[C
6 6H66]] over
66H
6 ] over
3 ] (b).
Figure
15. Poincaré
sections [C
over [NO]
6H
] over
[O3][O
(b).
(2)
Sustainability 2016, 8, 838
12 of 15
Qualitatively is provided that to a reduction of [C6 H6 ] under a certain threshold level corresponds
an increase of [O3 ]. Such behaviour could be attributed to exogenous state variable, which were not
taken into account in the present work.
4. Conclusions
Observance of air quality standards represents a great challenge in cities, especially in the ones in
which traffic and other additional sources are combined with bad weather conditions. With reference
to this issue, the results herein provided shown that the major pollutant concentrations are observed
in the winter seasons during the intense traffic flow and the threshold values are often exceeded.
Furthermore, according to the coefficient of variation values, Rome has a strongly unsteady behaviour
in terms of a family of pollutant concentration which fluctuate significantly.
[C6 H6 ] are strongly related to the concentration of pollutants species such as [CO], [NO] and
[NO2 ], as confirmed by the cross-correlation analysis. This correlation is due to the fact that nitrogen
oxide prompts and carbon oxides are combustion products and that their fluctuations are caused by
the oscillations of the city traffic flux. [O3 ] is weakly related to the production of benzene and it is also
in phase opposition with it. For this reason, the Poincaré section of [C6 H6 ] upon [NO] and [C6 H6 ]
upon [O3 ] was investigated. It is worth noticing that there is a strong linear dependence between
[C6 H6 ] and [NO] and a more complex interdependence of [O3 ] and [C6 H6 ]. Qualitatively is provided
that, to a reduction of [C6 H6 ] under a certain threshold level which corresponds to an increase of [O3 ].
Such behaviour could be attributed to exogenous state variable.
The main objective for the reduction of air pollutant is to reach, thanks to the implementation of
suitable mobility policies, an urban sustainable development, i.e., to improve traffic mobility conditions,
to increase road safety and to decrease traffic caused by pollution and to re-qualify urban spaces.
It includes rationalizing public space, safeguarding citizens’ health and life quality, and preserving
historical and architectural heritage.
The government is promoting both long and short-term activities to reach these goals. From the
analysis of the current air situation, it is clear that traffic is the reason why the pollutant concentrations
are so high. Traffic is the main source of [CO], [C6 H6 ] and [PM10 ] concentrations.
In order to reduce the pollution in Italian cities and in particular in Rome, measures are needed to
decrease the level of the various substances dispersed into the air. One of the main actions that can be
performed is the reduction or elimination of the use of the most polluting cars, i.e., cars Euro 0, 1 and
2. As a matter of fact, Roma Capitale has imposed the circulation reduction of these cars permanently
from 15 December 2015 [46].
Regarding the reduction of the pollutants due to buildings heating systems, it is necessary to
replace traditional boilers with more efficient systems, such as condensing boilers. As a matter of
fact, the condensing boilers allow reducing the utilization of combustion and a consequence decrease
of emission.
However, the boilers are only one element of the heating systems. Its efficiency depends on other
elements such as distribution, emission and regulation. Using condensing boilers coupled with other
high heating systems elements, can improve the total efficiency and reduce the environmental emission.
The Poincaré sections in Figure 15 gives a significant contribution to the modeling purpose:
such as dynamical model (ODE) or regression approach (LUR) (see among many [47,48]).
More specifically, the experimental analysis shows a correlation between [C6 H6 ] over [NO] and
[C6 H6 ] over [O3 ]. These aspects are well known for small scale zero-dimensional reactor, but for very
large scale problems as the city domain is not commonly investigated. Therefore, a reduction of a three
dimensional large scale process to zero-dimensional phenomenon as in homogeneous volume can be
consider the first step for a mathematical model development by means of an ODE system as follows:
x = f [ x (t)] where x (t) = {[NO] , [O3 ] , [C6 H6 ] , . . .}
(3)
Sustainability 2016, 8, 838
13 of 15
This aspect is the fundamental task to predict the time evolution of the pollutant species generated
within an urban domain.
Furthermore, as a result, it can be clearly concluded that Rome has a strongly unsteady behaviour
in terms of pollutant concentration. All-time series are positive skewed, indicating that some short
time rare events, of pollutant concentration, have an order of magnitude bigger than the expected
values. Such as intermittent behaviour must be taken into account for modelling purpose.
Acknowledgments: The authors are grateful to ARPA Lazio which provided the experimental data.
Author Contributions: The work was designed by Gabriele Battista, Tiziano Pagliaroli and Roberto De Lieto Vollaro.
Gabriele Battista and Tiziano Pagliaroli wrote the paper and perform the statistic and cross-statistic analysis.
English corrections were revised by Luca Mauri and Carmine Basilicata. Finally, Roberto De Lieto Vollaro
supervised the work related to the paper. All authors read and approved the final manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
International Agency for Research on Cancer (IARC). Outdoor air pollution. In IARC Monograph on the
Evaluation of Carcinogenic Risks to Humans; International Agency for Research on Cancer: Lyon, France, 2015;
Volume 109.
Mayer, H. Air pollution in cities. Atmos. Environ. 1999, 33, 4029–4037. [CrossRef]
Guerrieri, M.; Corriere, F.; Rizzo, G.; Casto, B.L.; Scaccianoce, G. Improving the Sustainability of
Transportation: Environmental and Functional Benefits of Right Turn by-Pass Lanes at Roundabouts.
Sustainability 2015, 7, 5838–5856. [CrossRef]
Schiavon, M.; Redivo, M.; Antonacci, G.; Rada, E.C.; Ragazzi, M.; Zardi, D.; Giovannini, L. Assessing the air
quality impact of nitrogen oxides and benzene from road traffic and domestic heating and the associated
cancer risk in an urban area of Verona (Italy). Atmos. Environ. 2015, 120, 234–243. [CrossRef]
Lim, S.S.; Vos, T.; Flaxman, A.D.; Danaei, G.; Shibuya, K.; Adair-Rohani, H.; Adair-Rohani, H.; Amann, M.;
Anderson, H.R.; Andrews, K.G.; et al. A comparative risk assessment of burden of disease and injury
attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: A systematic analysis for the
Global Burden of Disease Study 2010. Lancet 2013, 380, 2224–2260. [CrossRef]
Straif, K.; Cohen, A.; Samet, J. Air Pollution and Cancer. IARC Scientific Publication No. 161; International Agency
for Research on Cancer: Lyon, France, 2015.
Ghedini, N.; Gobbi, G.; Sabbioni, C.; Zappia, G. Determination of elemental and organic carbon on damaged
stone monuments. Atmos. Environ. 2000, 34, 4383–4391. [CrossRef]
Coppi, M.; Quintino, A.; Salata, F. Numerical study of a vertical channel heated from below to enhance
natural ventilation in a residential building. Int. J. Vent. 2013, 12, 41–49. [CrossRef]
Salata, F.; Golasi, I.; de Lieto Vollaro, E.; Bisegna, F.; Nardecchia, F.; Coppi, M.; Gugliermetti, F.;
de Lieto Vollaro, A. Evaluation of different urban microclimate mitigation strategies through a PMV analysis.
Sustainability 2015, 7, 9012–9030. [CrossRef]
De Lieto Vollaro, R.; Evangelisti, L.; Carnielo, E.; Battista, G.; Gori, P.; Guattari, C.; Fanchiotti, A. An Integrated
Approach for an Historical Buildings Energy Analysis in a Smart Cities Perspective. Energy Procedia 2014, 45,
372–378. [CrossRef]
Evangelisti, L.; Battista, G.; Guattari, C.; Basilicata, C.; de Lieto Vollaro, R. Influence of the Thermal Inertia in
the European Simplified Procedures for the Assessment of Buildings’ Energy Performance. Sustainability
2014, 6, 4514–4524. [CrossRef]
Salata, F.; de Lieto Vollaro, A.; de Lieto Vollaro, R.; Davoli, M. Plant reliability in hospital facilities.
Energy Procedia 2014, 45, 1195–1204. [CrossRef]
Evangelisti, L.; Battista, G.; Guattari, C.; Basilicata, C.; de Lieto Vollaro, R. Analysis of Two Models for
Evaluating the Energy Performance of Different Buildings. Sustainability 2014, 6, 5311–5321. [CrossRef]
De Lieto Vollaro, R.; Vallati, A.; Bottillo, S. Differents Methods to Estimate the Mean Radiant Temperature in
an Urban Canyon. Adv. Mater. Res. 2013, 650, 647–651. [CrossRef]
D’Orazio, A.; Fontana, L.; Salata, F. Experimental study of a semi-passive ventilation grille with a feedback
control system. Rev. Sci. Instrum. 2011, 82, 085107. [CrossRef] [PubMed]
Sustainability 2016, 8, 838
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
14 of 15
Coppi, M.; Quintino, A.; Salata, F. Fluid dynamic feasibility study of solar chimney in residential buildings.
Int. J. Heat Technol. 2011, 29, 1–5.
Salata, F.; Golasi, I.; de Lieto Vollaro, R.; de Lieto Vollaro, A. Outdoor thermal comfort in the Mediterranean
area. A transversal study in Rome, Italy. Build. Environ. 2016, 96, 46–61. [CrossRef]
Salata, F.; Alippi, C.; Tarsitano, A.; Golasi, I.; Coppi, M. A first approach to natural thermoventilation of
residential buildings through ventilation chimneys supplied by solar ponds. Sustainability 2015, 7, 9649–9663.
[CrossRef]
Vallati, A.; De Lieto Vollaro, A.; Golasi, I.; Barchiesi, E.; Caranese, C. On the impact of urban micro climate
on the energy consumption of buildings. Energy Procedia 2015, 82, 506–511. [CrossRef]
Battista, G.; Carnielo, E.; Evangelisti, L.; Frascarolo, M.; de Lieto Vollaro, R. Energy Performance and
Thermal Comfort of a High Efficiency House: RhOME for denCity, Winner of Solar Decathlon Europe 2014.
Sustainability 2015, 7, 9681–9695. [CrossRef]
Memon, R.A.; Leung, D.Y.C.; Liu, C. An investigation of urban heat island intensity (UHII) as an indicator of
urban heating. Atmos. Res. 2009, 94, 491–500. [CrossRef]
Goia, F.; Zinzi, M.; Carnielo, E.; Serra, V. Characterization of the optical properties of a PCM glazing system.
Energy Procedia 2012, 30, 428–437. [CrossRef]
Paolini, R.; Zinzi, M.; Poli, T.; Carnielo, E.; Mainini, A.G. Effect of ageing on solar spectral reflectance
of roofing membranes: Natural exposure in Roma and Milano and the impact on the energy needs of
commercial buildings. Energy Build. 2014, 84, 333–343. [CrossRef]
Goia, F.; Zinzi, M.; Carnielo, E.; Serra, V. Spectral and angular solar properties of a PCM-filled double glazing
unit. Energy Build. 2015, 87, 302–312. [CrossRef]
Zinzi, M.; Carnielo, E.; Federici, A. Preliminary studies of a cool roofs’ energy-rating system in Italy.
Adv. Build. Energy Res. 2014, 8, 84–96. [CrossRef]
Carnielo, E.; Zinzi, M. Optical and thermal characterisation of cool asphalts to mitigate urban temperatures
and building cooling demand. Build. Environ. 2013, 60, 56–65. [CrossRef]
De Lieto Vollaro, A.; Galli, G.; Vallati, A.; Romagnoli, R. Analysis of thermal field within an urban canyon
with variable thermophysical characteristics of the building’s walls. J. Phys. Conf. Ser. 2015, 655, 012056.
[CrossRef]
De Lieto Vollaro, A.; Galli, G.; Vallati, A. CFD Analysis of Convective Heat Transfer Coefficient on External
Surfaces of Buildings. Sustainability 2015, 7, 9088–9099. [CrossRef]
Chan, A.T.; So, E.S.P.; Samad, S.C. Strategic guidelines for street canyon geometry to achieve sustainable
street air quality. Atmos. Environ. 2001, 35, 4089–4098. [CrossRef]
Oke, T.R. Street design and urban canopy layer climate. Energy Build. 1988, 11, 103–111. [CrossRef]
Britter, R.E.; Hanna, S.R. Flow and dispersion in urban areas. Annu. Rev. Fluid Mech. 2003, 35, 469–496.
[CrossRef]
Fisher, R.A.; Tippett, L.H.C. Limiting forms of the frequency distribution of the largest or smallest member
of a sample. Math. Proc. Camb. Philos. Soc. 1928, 24, 180–190. [CrossRef]
Martins, E.S.; Stedinger, J.R. Generalized maximum-likelihood generalized extreme-value quantile estimators
for hydrologic data. Water Resour. Res. 2000, 36, 737–744. [CrossRef]
Kütchenhoff, H.; Thamerus, M. Extreme value analysis of Munich air pollution data. Environ. Ecol. Stat.
1996, 3, 127–141. [CrossRef]
Smith, R.L. Extreme Value Analysis of Environmental Time Series: An Application to Trend Detection in
Ground-Level Ozone. Stat. Sci. 1989, 4, 367–377. [CrossRef]
Rao, S.T.; Sistla, G.; Henry, R. Statistical analysis of trends in urban ozone air quality. J. Air Waste Manag. Assoc.
1992, 42, 1204–1211. [CrossRef]
Nor, A.M.A.; Mohd, B.A.; Ahmad, Z.A. Bayesian Extreme for Modeling High PM10 Concentration in Johor.
Procedia Environ. Sci. 2015, 30, 309–314.
Shi, L.; Stuart, B.; Feng-Chiao, S.; Bhramar, M. Addressing extrema and censoring in pollutant and exposure
data using mixture of normal distributions. Atmos. Environ. 2013, 77, 464–473.
Shen, L.; Mickley, L.J.; Gilleland, E. Impact of increasing heat waves on US ozone episodes in the 2050s:
Results from a multimodel analysis using extreme value theory. Geophys. Res. Lett. 2016, 43, 4017–4025.
[CrossRef] [PubMed]
Sustainability 2016, 8, 838
40.
41.
42.
43.
44.
45.
46.
47.
48.
15 of 15
Roma Capitale, Piano Generale del Traffico Urbano (in English: City Urban Traffic General Plan).
April 2015. Available online: https://www.comune.roma.it/pcr/it/dip_mob_delibere.page (accessed on
22 August 2016).
United States Environmental Protection Agency (EPA). Air Quality Planning and Standards. Available online:
https://www3.epa.gov/airquality/ (accessed on 22 August 2016).
European Commission. Available online: http://ec.europa.eu/ (accessed on 22 August 2016).
European Union. Directive 2008/50/EC of the European Parliament and of the Council of 21 May 2008 on
Ambient Air Quality and Cleaner Air for Europe. Available online: http://eur-lex.europa.eu/LexUriServ/
LexUriServ.do?uri=OJ:L:2008:152:0001:0044:en:PDF (accessed on 22 August 2016).
ARPA Lazio—Centro Regionale della Qualità dell’Aria (in English: Regional Center of Air Quality).
Available online: http://www.arpalazio.net/ (accessed on 22 August 2016).
ARPA Lazio. Valutazione della Qualità dell’aria—2015 (In English: Air Quality Assessment—2015). RT/DAI/16/01;
ARPA Lazio: Rieti, Italy, 2015.
Roma Capitale, Fascia Verde e Anello Ferroviario (in English: Green Band and Ring Rail). Available online:
http://www.agenziamobilita.roma.it/it/servizi/orari-ztl/fascia-verde-anello-ferroviario.html (accessed on
22 August 2016).
Beelen, R.; Hoek, G.; Vienneau, D.; Eeftens, M.; Dimakopoulou, K.; Pedeli, X.; Tsai, M.; Künzli, N.;
Schikowski, T.; Marcon, A.; et al. Development of NO2 and NOx land use regression models for estimating
air pollution exposure in 36 study areas in Europe—The ESCAPE project. Atmos. Environ. 2013, 72, 10–23.
[CrossRef]
Eeftens, M.; Beelen, R.; de Hoogh, K.; Bellander, T.; Cesaroni, G.; Cirach, M.; Declercq, C.; Dėdelė, A.;
Dons, E.; de Nazelle, A.; et al. Development of land use regression models for PM2.5, PM2.5 absorbance,
PM10 and PMcoarse in 20 European study areas; results of the ESCAPE project. Environ. Sci. Technol. 2012,
46, 11195–11205. [CrossRef] [PubMed]
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).