Machine Translated by Google

Urban Climate 52 (2023) 101739

Contents lists available at ScienceDirect

Urban Climate
journal homepage: www.elsevier.com/locate/uclim

Concentration levels of atmospheric contaminants in Brazilian

cities measured by passive sampling
b
Franciele O. Campos da Rocha a,*, Vˆ ania P. Campos , Gisele O. da Rocha b,c,
Marcos A. Bezerra d, Anne Valesca S. Brito b , Indiara dos Santos Sampaio b
to
Universidade Federal de Roraima, UFRR, Center of Sciences and Technology, Department of Chemistry, 69301-160 Boa Vista, RR, Brazil b

Universidade Federal da Bahia, UFBA, Institute of Chemistry, Campus Ondina, 40170-270 Salvador, BA, Brazil
c
Interdisciplinary Center in Energy and Environment, CIEnAm, UFBA, CEP 40170-115 Salvador, BA, Brazil
d
Universidade Estadual do Sudoeste da Bahia, Department of Sciences and Technologies, Jequi´e, Bahia 445205-490, Brazil

ARTICLE INFO ABSTRACT

Keywords: Measurements of air pollutants were made using passive sampling in Brazilian urban centers. The levels of SO2 and NO2
Urban atmospheric pollution exceeded the WHO recommendations in S˜ ao Paulo (SP), Londrina and Rio de Janeiro. Formaldehyde concentrations
Vehicle emissions
exceeded international recommendations in all urban centers and benzene concentrations exceeded 1.7 ÿg mÿ level
Chemometrics 3
,
associated with the probability of leukemia, according to the WHO. Sewage discharge in the Pinheiros River - SP and in
Air pollution
Lake Pampulha - BH contributed to higher levels of H2S and NH3 in these urban centers. The BH at-mosphere was also
Urban air quality
influenced by emissions from landfills. Organic and inorganic acids contribute similarly to atmospheric acidity. 71% of the
calculated ratios between concentrations of formic and acetic acids were < 1, demonstrating the predominant influence
of vehicle emissions on the atmospheric levels of these acids. Multivariate analysis showed that VOCs and organic acids
are more influenced by temperature and solar radiation. The stronger direct relationship of NOx was with wind direction,
showing competition between local emission and photochemical aging of these compounds brought from other locations.
On a global scale, studies of this type are necessary and important to increase the accuracy of exposure estimates in
different places where people live.

1. Introduction

The world's urban population has grown rapidly in recent years and, according to the United Nations, it is expected that urban centers
will house 70% of the world's population by 2050. Brazil is among the 7 countries whose expansion represents 50% of this population
( UN-Habitat, 2022). The intensification of urbanization processes results in fossil fuel consumption, deforestation, burning, waste
generation and, consequently, emissions of trace substances such as gases and particulate matter that degrade air quality, making urban
air pollution a global public health concern (Santana et al. , 2019; Liu et al., 2022).
It is estimated that 4.2 million premature deaths worldwide are linked to environmental air pollution, mainly heart disease, stroke,
chronic obstructive pulmonary disease, lung cancer and acute respiratory infections in children. The impact of outdoor air pollution falls
disproportionately on people in low- and middle-income nations, where 89% of premature deaths occur from ambient air

* Corresponding author.
E-mail address: [email protected] (FO Campos da Rocha).

https://doi.org/10.1016/j.uclim.2023.101739 Received 14
April 2023; Received in revised form 2 September 2023; Accepted 22 October 2023 Available online 29 October 2023

2212-0955/© 2023 Elsevier BV All rights reserved.

Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

pollution (WHO, 2022).

The atmospheric chemical behavior of trace substances is very complex, so that the degradation of air quality results from a set of
factors such as the type of emissions and intensity of fixed or stationary sources, atmospheric reactions, topography and climatic
conditions, which determine the occurrence of adverse effects of air pollution on its receptors (Seinfeld and Pandis, 2016; Zscheischler et
al., 2020). The substances that are emitted are frequently transformed into more toxic substances than their precursors, resulting in
greater damage to the fauna, flora and humans in the places where they are deposited.
People's concern about the damage caused by pollutants continuously emitted into the environment has increased the pressure for
regulatory measures to be implemented with a view to constantly assessing environmental quality and its preservation. In Brazil, despite
the recent update on CONAMA resolution 003/90 to resolution 491/2018 (Brazil, 2018), it is observed that there are still few legislated
gaseous pollutants (O3, NO2, SO2 and CO), compared to the number of contaminants proven to be toxic and harmful to humans, fauna
and flora, such as aldehydes, BTEX, organic and inorganic acids, besides NH3.
Furthermore, air quality monitoring in the country is still restricted and unsatisfactory in terms of territorial coverage, number of
monitored parameters and representativeness in measurements, due to management difficulties and the low number of technicians
available, in addition to the lack of resources for the purchase and maintenance of equipment and composition of monitoring networks.
Until 2017, there were only 24 Brazilian cities with some air quality monitoring, all located in the Southeast region of Brazil, and only 8 of
them monitored all conventional pollutants (Andreao ˜ et al., 2018).
The most widespread and applied atmospheric monitoring technique is active sampling, mainly through the use of continuous monitors
that consist of complicated and expensive technology (Rocha and Campos, 2023). It is necessary to develop alternative strategies that
are economically viable, easy to operate and that provide reliable information on the concentrations of the monitored pollutants.

Passive sampling has been shown to be a viable alternative for atmospheric monitoring. Passive samplers are devices capable of
trapping atmospheric gases or vapors at a rate controlled by a physical process, such as diffusion or permeation, without involving the
active movement of air through the sampler. They do not require a suction pump to force air to be sampled; In addition to simplicity and
low cost, they require little maintenance, do not depend on air flow calibration and are easy to operate. This type of device has been widely
used to sample gases and vapors in the atmosphere (Campos et al., 2010; Fantozzi et al., 2015; Yli-Pelkonena et al., 2017; Santana et
al., 2017; Cruz et al. ., 2020; Cruz et al., 2022; da Rocha et al. , 2022; Rocha and Campos, 2023), including at official atmospheric
monitoring stations in Germany (SMVKU, 2022). Recently, it has also been used to sample ultra-fine air pollutants in the atmosphere
(Neckel et al., 2023).
Studies of the atmosphere of large Brazilian urban centers can be found in the literature but, in general, there is still much to be done.
Furthermore, it is known that vehicle emissions have been recognized as one of the most important anthropogenic sources of gas and
particulate emissions, including toxic species. However, very little has been studied and there is relatively little information about the real
contribution of gases and particles emitted by vehicles in Brazil.
Therefore, the seasonal assessment of atmospheric contaminants helps in the understanding of the main processes that may be
involved in the variation of pollutant concentrations in the air and their respective health risks. In this study, concentrations of 16 gaseous
pollutants/contaminants (SO2, O3, NO2, NOx, NO, NH3, BTEX, HCOH, CH3COH, HCl, HNO3, HCOOH and CH3COOH) were evaluated
in 5 large Brazilian urban centers (Sao ˜ Paulo ,Salvador, Rio de Janeiro, Londrina and Rio de Janeiro), using passive sampling and data
correlation through chemometric treatment, including the variation of meteorological conditions and investigation of the impact of vehicle
emissions in these centers.
In relation to the main effects on human health and on the quality of the environment related to the pollutant gases considered in this
study, we can mention: (a) Human carcinogenic effect: benzene, ethylbenzene, acetaldehyde and formaldehyde (also mutagenic)
(IARC, 1987; WHO, 2000; Kro and Namies, 2010; USEPA, 2022a, 2022b); (b) Breathing problems affecting lung bronchi and alveoli: O3,
SO2, NO2, NH3 and H2S (WHO, 2000; Shi et al., 2015; USEPA, 2017a, 2017b; Wang and Thomas, 2017; Conforti et al., 2018 ; NIOSH,
2019; Padappayil and Borger, 2023); (c) Participation in the formation of acid rain: inorganic acids (HCl and HNO3), organic acids (HCOOH
and CH3COOH) and SO2 (Nolte et al., 1997; NPI, 2014; Stern, 2015; USEPA, 2017a); (d) Precursors in the formation of secondary
pollutants: O3, NO and CH3COH; CH3COH is an important precursor of secondary pollutants that are highly irritating to the eyes and
phytotoxic: peroxyacyl nitrates (PANs) and peroxybenzyl nitrates (PBNs) (WHO, 2000; IARC, 2004; Muttray et al., 2009; NPT, 2011;
USEPA , 2017a; USEPA, 2022b). On the basis of the above-mentioned references, many other effects are still known for these compounds.

According to the WHO (2021), air quality management for the assessment of trends and to estimate exposure for epidemiological
analyzes is traditionally based on measurements of air pollutant concentrations in official monitoring networks. The WHO considers that,
despite the growth in the number of global monitoring sites, which are frequently restricted to large cities, coverage is inadequate to
accurately estimate exposure in the different places where people live. This occurs even for the most commonly monitored pollutants,
such as the classic ones contemplated in this study (NO2, SO2 and O3), among other important atmospheric pollutants measured in
urban centers with different characteristics. Thus, studies of this type are of paramount importance on a global scale.

2. Materials and methods

2.1. Sampling

2.1.1. Study area

Fig. 1. locates the five large Brazilian urban centers analyzed (S˜ ao Paulo, Rio de Janeiro, Salvador, Belo Horizonte and Londrina.

2
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Table 1 shows the details of these points.

2.1.2. Description of the sampling methodology

AnalyzeAr kits, a low-cost system for assessing air quality, were used to sample the 16 gases; each kit comprised 15 passive samplers
(PSs) (Rocha and Campos, 2023). Sampling was carried out passively in triplicate, using 2 kits in each campaign, one for an exposure
period of 7 days, relative to SO2, O3, NO2, NOx and NH3 measurements, and another for an exposure period of 14 days for mea-
surements of BTEX, HCOH, CH3COH, HCl, HNO3, HCOOH and CH3COOH. The kits were exposed at a minimum height of 2 m from
the ground. Fig. 2 shows the drawing of the AnaliseAr kit with the distribution of the passive samplers (PS) for exposure, a photo of the
exposed kits and a detailed description of the PS model. Six campaigns were carried out between August/2017 and June/2018. After
each sampling period, the kits were returned to the laboratory for analysis of the compounds fixed in the PSs.

2.2. Analytical methodology for preparing and analyzing passive samplers

The kits containing the passive samplers (PSs) were assembled at the Environmental Analytical Chemistry Laboratory (LAQUAM).
After cleaning, filter impregnation and assembling, the samplers were sealed with the protective/carrier box and Parafilm® until sampling
began.
After exposure of the PSs, the analytes fixed on the impregnated filters were extracted in an ultrasonic bath without heating, followed
by centrifugation. The analytical methodology and sampling are detailed in the kit technical note (Rocha and Campos, 2023).
For the impregnation solutions of the filters to be placed in the PSs, the following reagents were used: Na2CO3, in PSs for SO2, O3, HCl
and HNO3; TEA (Triethanolamine) for CH3COOH, HCOOH and NO2; TEA + PTIO (2-phenyl-4,4,5,5-tetramethylimidazoline-1-oxyl-3-
oxide), for NOx; H2C2O4 (oxalic acid), for NH3; zinc acetate + NaOH + trisodium citrate, for H2S; 2,4-DNPH (dinitrophenylhydrazine), for
CH3CHO and HCHO and the adsorbent activated carbon in the PS for BTEX. The extractions of the compounds fixed in the PSs were
aqueous, except for aldehydes, extracted with acetonitrile, and BTEX compounds, with carbon tetrachloride. For the analysis, chro-
matographic analytical techniques were used, in addition to UV–Vis molecular absorption spectrophotometry (Griess Saltzman, indophenol
blue and methylene blue).

Fig. 1. Location of the sampling points in the urban centers considered in the study.

3
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Table 1
Details of sampling sites.

Sao Paulo

City Sao
˜ Paulo (Megacity)
Number of inhabitants* 12,252,023
Vehicle fleet** 8,341,669
Location External area of the Institute of Astronomy, Geophysics and Atmospheric Sciences (IAG) – 226, Mat˜ ao Street, Cidade Universitaria 23.5602107ÿ S /
´
46.7346811ÿ W
Coordinators Close to the Tietˆe and Pinheiros Marginal
Main characteristics of the (intense vehicular flow); industrial influence
sampling site

Bay
City Salvador
Number of inhabitants* 2,872,347
Vehicle fleet** 920,604
Terrace (~ 10 m from the ground) of the Interdisciplinary Center for Energy and Environment (CIENAM) – S/N, Bar˜ ao de Jeremoabo Street, Ondina
Location
Campus 13.000193ÿ S / 38.507263ÿ W
Coordinators
Main characteristics of the
UFBA Campus, Coastal area; close to the city's main thoroughfares; high afforestation;
sampling site

Minas Gerais
City Belo Horizonte
Number of inhabitants* 2,512,070
Vehicle fleet** 2,088,132
ˆ
Location Entrance of the Federal University of Minas Gerais – 6627, Presidente Antonio Carlos Avenue 19.8715335ÿ S /
Coordinators 43.9551401ÿ W Intense vehicle flow
Main characteristics of the Border with Fern˜ ao
sampling site Dias Highway; close to the Municipal Sanitary Landfill and Macaúbas waste treatment center

Paran´ a
City Londrina
Number of inhabitants* 569,733
Vehicle fleet** 399,434
Location 683, Castelo Branco Avenue
Coordinators 23.3186317ÿ S / 51.1933064ÿ W
Main characteristics of the
Next to Celso Garcia Cid Highway, with intense vehicle flow
sampling site

Rio de Janeiro
City Rio de Janeiro
Number of inhabitants* 6,718,903
Vehicle fleet** 2,839,512
Location Terrace of the UFRJ Technology Center (7th floor), 149, Athos da Silveira Ramos Avenue 22.8598541ÿ S /
Coordinators 43.3391842ÿ W
Main characteristics of the
Coastal area with industrial influence and intense vehicle flow
sampling site

* (IBGE, 2021).
** (SENATRAN, 2020).

2.3. Statistical analysis

2.3.1. Linear correlation matrix

Pearson correlation was used to analyze the correlation between numerical variables, offering measures of the bivariate
association (strength) of the degree of relationship between the variables, indicating the intensity and direction of the linear or non-
linear relationship between two variables 16.
According to Dancey and Reidy (2006), the correlation coefficient (r) can be qualitatively evaluated considering the following
ranges: r = 0.10–0.30, weak correlation; r = 0.40–0.6, moderate correlation and r = 0.70–1.0, strong correlation. To check whether
Pearson linear correlation is significant or not at 5% probability, the tabulated critical value of r was used for 18 data pairs: rcrit = |
0.50|.

2.3.2. Multivariate data analysis

The average concentration data of the 16 pollutants in the atmosphere of the 5 urban centers for the 6 sampling campaigns were
submitted to multivariate analysis, with application of principal component analysis (PCA) for the seasonal chemometric study
associated with meteorological parameters such as rainfall, relative humidity, temperature, solar radiation, wind speed and direction,

4
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Fig. 2. (a) Distribution of passive samplers (PS) in kits according to exposure time; (b) Representation of the exposure of the sampling system.
(c) detailed description of the PS model: (I) air inlet, (II) stainless steel mesh (III) Teflon membrane, (IV) diffusion space, (V) impregnated filter
(for BTEX a 2 mm high gap at the sampler bottom is limited by another stainless-steel screen and filled with activated charcoal, (VI) sampler bottom.

provided by the National Institute of Meteorology (INMET) (INMET, 2018). The STATISTICA 7.0 software was used with pre-processed data by autoscaling.

3. Results and discussion

The 6 field campaigns resulted in a total of 180 samples for the 16 analytes.
The results presented below were separated into groups according to the classification of pollutants/contaminants (Fig. 3).

3.1. Concentration of conventional air pollutants in Brazil: NO2, O3 and SO2

Among the studied gases, only NO2, SO2 and O3 are legislated in Brazil (Brazil, 2018). Mean concentrations ranged from 1.4 to 60 3
3 3
, 1.4 to 50 ÿg mÿ and 5.4 to 139 ÿg mÿ for NO2, SO2 and O3, respectively. Fig. 3(a) shows the variation in the average ÿg mÿ concentrations of these
compounds for the exposure period and their relationship with rainfall in the period. Table S1 presents the mean concentrations obtained for these pollutants and the
quantification limits of PSs.
SO2 was the only pollutant with average concentrations above those recommended by CONAMA resolution 419/2018 (Brazil, 2018) (40 ÿg mÿ ˜ Paulo, Londrina
3
); this
and Rio de Janeiro, and limit
in the 4thwas
andexceeded in the 3rd
6th campaigns sampling
again in Sao campaign in Saoof˜ Paulo.
the atmosphere For presented a concentration level very close to this value in the 3rd
Sao ˜ Paulo
3
campaign (139 ÿg mÿ O3, despite not exceeding the national limit of 140 ÿg mÿ ,
3
).
Regarding NO2, the concentrations exceeded the annual limit recently established by the WHO (WHO, 2021) of 10 ÿg mÿ campaigns in Sao 3rd campaign3 in all
in Londrina. NO2 ˜ Paulo, with levels up to 6 times higher than the recommendation, in 3 of the 4 campaigns in Rio de Janeiro and in the ˜ Paulo were similar to those
concentrations in Sao million inhabitants in Ganga basin, India. found by Behera et al. (2015) in Kanpur, city with >4

Vehicle emissions are currently identified as the main anthropogenic sources in urban areas for primary emissions of SO2 and NO2 and, consequently, for the formation of
tropospheric O3. In general, the highest values for these conventional pollutants were found in Sao ˜ Paulo - SP, which has the largest vehicle fleet, with the highest consumption
ofdiesel and type C gasoline in the country as well (Brazil, 2019a). Furthermore, it is the most industrialized among the studied centers, which justifies the higher concentrations

observed.

5
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Fig. 3. Average levels of the 16 gaseous atmospheric pollutants/contaminants in Brazilian urban centers between August/2017 and June/2018.

Salvador was the city with the lowest average concentrations, with the second smallest vehicle fleet among the studied centers, further corroborated by the
constant influence of cleaner air mass trajectories, predominating over the ocean, as shown by the reverse trajectories obtained using the NOAA HYSPLIT model
(NOAA, 2023) (Fig. S1).
Considering that Rio de Janeiro is the second center with the largest vehicle fleet and industrial development among those studied, higher levels of atmospheric
concentrations were expected. However, the concentrations found there at a level similar to urban centers such as Belo Horizonte and Londrina, which can be
explained by the influence of air masses mostly originating from the ocean (Fig. S1).
Londrina has the smallest vehicle fleet, when compared to the other studied centers. However, it presented higher average at-mospheric concentrations than
Salvador and Rio de Janeiro, even exceeding the legislated limits for SO2. This can be explained by the proximity of the sampling site to a highway with an intense
flow of heavy vehicles, in addition to light flex vehicles, characteristic of urban centers. On the other hand, the reverse trajectories of air masses that landed there in
the sampling periods, indicated pre-dominance over the continent, passing over urban areas (Fig. S1).

The concentration of pollutants in the atmosphere is directly influenced by the distribution and intensity of emissions of atmospheric pollutants, topography and
meteorological conditions. When evaluating the behavior of the pollutants studied over the sampling period, it is necessary to consider that, in the years in which the
samplings were carried out (2017/2018), the water cooling conditions of the Equatorial Pacific Ocean occurred, signaling the development of rainfall since August
2017 (beginning of sampling), which may interfere with the rainfall pattern in Brazil. Even being considered of low intensity, these conditions may have influenced the
variation of pollutants. More continuous rain in the Midwest and Southeast Regions, above-normal rainfall rates in the North and Northeast Regions, and irregularities
in the distribution in the South Region are expected in Brazil as average rainfall conditions in La years. As for temperature, values above normal are expected in the
South Region, normal to below normal in the Southeast and
˜
Little girl

Midwest Regions, and within the normal range in the Northeast and North Regions of the country (INMET, 2018).
˜
Rainfall influences the reduction in concentrations of pollutants from the atmosphere. The La Nina may have interfered with the average rainfall expected for the
periods and centers studied. Salvador, which normally has a rainy season from April to August, had

6
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

high rainfall rates in January/2018, a period characterized by drought in the region, which may have reduced the concentration levels in that period.

The characteristic dry periods for S˜ ao Paulo and Belo Horizonte (C1 and C6, Fig. 3) would be the most unfavorable for the dispersion of primary
pollutants due to low rainfall; However, it can be observed that, in 2017, there were two long drought periods, one of them from the end of August to
the end of September (C1 and C2), reinforcing the higher concentrations for these centers shortly after this period, caused for consecutive days without
rain, as well as the high incidence of solar radiation and high temperatures.
Although rainfall in the following months was, most of the time, higher than the climatological averages, these conditions were not sufficient to
prevent an increase in the levels of these pollutants. November/2017 stands out presenting, in general, higher concen-trations of SO2, NO2 and O3,
even though it is a period with more rainfall and high temperatures, which could indicate wet removal of these pollutants and NO2 consumption by
photochemical reactions for ozone formation, on the hottest days with the highest incidence of solar radiation at the top of the atmosphere. Therefore,
the results suggest an increase in vehicle emissions of SO2 and NO2 in all cities for this month, overcoming the influence of wet rainfall and higher
temperatures. This can be explained by the increase in transport charge in the month before Christmas.

The concentration levels found in this study for SO2, NO2 and O3 are similar to other studies in urban areas (Campos et al., 2006; Di Carlo et al.,
2014; Mraihi et al., 2015; Abe and Miraglia, 2016; Omidvarborna et al., 2018; Schuch et al., 2019; Squizzato et al., 2021; Jion et al., 2023) and to
concentration levels found by state government agencies responsible for monitoring the centers considered in this study, with the exception of Salvador,
where quality was not being monitored during the period.

3.2. Nitrogenous pollutants/ contaminants (NOx, NH3, HNO3)

3 3 3
Mean concentrations ranged from 1.9 to 64 ÿg mÿ for NOx, 1.6 to 34 ÿg mÿ for NH3, and < 0.76 to 18 ÿg mÿ for HNO3. From the
3
difference between the values obtained for NOx and NO2, mean NO concentrations were calculated, ranging from 0.32 to 19 ÿg mÿ Fig. 3(b) shows .
the variation in the average concentrations of these pollutants as a function of the sampling period (dry and rainy).
Table S2 presents the average concentrations obtained for these pollutants, with the respective quantification limits of the passive samplers. ˜ Paulo
and Belo
Sao
Horizonte presented NOx levels higher than the recommended by the European Union (European Union, 2008) (30 ÿg mÿ Berlin (SMVKU, 2022).
3
The results ) in 4 of the 6 campaigns carried out. This was also reported in the official monitoring stations in Lisbon (QualAR, 2021) and
for nitrogen oxides show a direct correlation with diesel consumption, the predominant source of emission of these gases in urban atmospheres. S˜ ao
Paulo, the center with the highest consumption of fossil fuels and industrial development, again presented the highest concentration values for these
nitrogenous pollutants/pollutants.
Although rainfall was, most of the time, higher than the climatological averages in S˜ ao Paulo and Belo Horizonte, these conditions were not
sufficient to avoid episodes of high concentrations of nitrogen oxides indicating, in this case, a continuous source of NO2, whose influence on
atmospheric concentration levels is greater than the influence of the wet deposition, which would tend to decrease them. Despite a small variation in
the average temperature between the periods and locations studied, it is also observed that the highest average concentrations of nitrogen oxides
were reported in warmer months with a higher incidence of solar irradiation.
In urban areas, house fires, traffic and industrial activity are generally considered the main sources of NH3 (Sun et al., 2017; Teng et al., 2017).
NH3 concentrations were higher between January and June/2018 for all urban centers which, according to INMET meteorological data, indicate a
period of greater rainfall and higher temperatures (Fig. 3(b)). This fact suggests that rainfall and temperature were not determining factors for NH3
concentrations in the atmosphere, as much as the intensity of anthropogenic emissions from the use of biofuels, industrial sectors or the use of
fertilizers in agriculture could have been. This can be confirmed by observing the predominance of air mass trajectories over the continent (Fig. S1)
which, in addition to the characteristic vehicular emissions, are associated with those coming from industrialized areas, such as S˜ ao Bernardo do
Campo and Osasco in Sao ˜ Paulo and/or agricultural areas as is the case of Belo Horizonte, capital of the largest producer and consumer of fertilizers
among the other locations considered in this study and also the 6th largest agricultural producer in the country (Brazil, 2019b).

In the sampling periods with a significant increase in the concentration of NH3 in Salvador-BA, C3 and C4, the air masses that arrived at the
sampling point were also influenced by emissions from the Industrial Pole of Camaçari – BA, located 50 km from Sal -vador (Fig. S2), where >90
chemical industries operate, some with intense fertilizer production.
HNO3 is considered an important product of the oxidation of gaseous nitrogenous compounds and an important contributor to acid rain (Hassan et
al., 2012; Anglada et al., 2020; Nojiri et al., 2022). The concentrations of HNO3 were observed in Salvador during the dry period and in S˜ ao Paulo,
Rio de Janeiro and Londrina during the rainy period, where, despite intense rainfall, temperatures and solar irradiation were the highest, contributing
to the photochemical formation of HNO3 from gaseous precursors (HO* + NOx ÿ HNO3). The seasonal variation of this acid was more pronounced
than that of NH3 and NOx, which can be explained by the variation in photochemical activity and gas-particle transformations (HNO3 + NH3 ÿ NH4NO3)
(Hassan et al., 2012; Seinfeld and Pandis, 2016).

Similar to NO2, previously discussed, HNO3 concentrations in Belo Horizonte are higher in the rainy season, which corroborates the
continuous source justification of its precursor (Fig. 3b).
Urban centers whose air mass trajectories were influenced by marine aerosol had lower levels of HNO3, indicating that this acid reacts favorably
with NaCl from marine spray, producing particulate matter and gaseous HCl, which contributes to the irreversible formation of fine nitrate particles (Eq.
( 1)).

HNO3 + NaClÿNaNO3(part.) + HCl(g) (1)

7
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Concentration levels for NOx, NH3, HNO3 are in the same order of magnitude as other recent studies in urban areas (Cyrys et al.,
2012; Kim et al., 2012; Agudelo-Castaneda ˜ et al., 2013; Vieira-Filho et al., 2016; Alfoldy et al., 2019).

3.3. VOCs: BTEX, formaldehyde and acetaldehyde

Mean concentrations of BTEX compounds ranged from <0.94–29, <0.80–126, <1.0–2.3, <0.91–2.5, <0.90–1.7 ÿg mÿ zene, toluene, ethylbenzene, o-3 for ben-
xylene and m,p-xylene, respectively; for formaldehyde and acetaldehyde, the mean concentrations ranged from 5.1 to 17 and 1.3–17 ÿg mÿ pollutants as a function
3
of the sampling period (dry and rainy). Table S3 , respectively. Fig. 3(c), (d) and (e) shows the variation in the average concentrations of these
presents the average concentrations obtained for these pollutants with the respective quantification limits of the passive samplers.

In Brazil, there is no legislation for these compounds. However, considering international recommendations, it is observed that, in general, benzene presented
values higher than those associated with the probability of leukemia of 1: 100,000 people, according to the World Health Organization (WHO) (> 1.7 ÿg mÿ standard
3
of 5.0 ÿg mÿ Benzene also showed higher concentrations in ) (WHO, 2000). In S˜ ao Paulo, the benzene concentration level exceeded the annual
3
Sao , established by the European Union through Directive 2008/50/EC (European Union, 2008). ˜ Paulo in the dry
period, suggesting intense local sources, such as vehicle and industrial emissions for VOCs, in
addition to the predominance of air mass trajectories through urban areas. The other cities with less influence from industrial sources had a lower concentration of
this compound.
Toluene was the pollutant among the sampled VOCs with the highest concentration levels. Its average concentration in Belo Horizonte was 3 to 20 times the
average found in the other urban centers considered in this study, indicating an intense local source.
Information on the trajectory of reverse air masses suggests that the sampling site, located within a 10 km radius of the landfill and waste treatment center in
Macaúbas, is influenced by toluene emissions in biogas, which is among the most common VOCs in landfill emissions, favored by the direct volatilization of sources
such as foams, solvents and paints (Fang et al., 2012; Liu et al., 2016; de S´ a Borba et al., 2017; Manco et al., 2022).

3 3
The concentrations of acetaldehyde (< 1.3 ÿg mÿ in SSA – 17 ÿg mÿ in SP - Table S3) were below the reference value established
3
by the ESL – Effects Screening Levels – Texas Commission on Environmental Quality (TCEQ, 2016), of 45 ÿg mÿ formaldehyde, all concentrations. However, for in
3 3
(5.3 ÿg mÿ in SSA – 17 ÿg mÿ BH - Table S3) exceeded the maximum value recommended (3.3 ÿg). It is observed
3
mÿ that the ratio between the concentrations of formaldehyde and acetaldehyde are lower than those found in the past (Grosjean et al., 2002; Correa et al., 2010;
Rodrigues et al., 2012; Nogueira et al., 2014; Santana et al., 2017), probably due to the decrease in the use of ethanol in the country; a smaller amount of
acetaldehyde is now emitted into the atmosphere.
In Salvador, the VOCs that showed greater seasonal variation were acetaldehyde and toluene, with higher concentrations in the period of lower solar irradiation,
indicating less photochemical removal and suggesting that direct emissions influenced more than the in situ formation of these compounds in the period. The
concentrations of xylenes decreased to a greater extent in warmer periods with more intense irradiation, when compared to other BTEX, which may be associated
with their higher reaction rate with HO* radicals (Atkinson, 1994), removing them more quickly from the troposphere. In addition, Salvador, as well as Rio de Janeiro,
are the cities with the highest solar irradiation throughout the year among the urban centers studied, favoring the photochemical removal of VOCs (Atkinson, 2000).
On the other hand, the atmosphere sampled in these two locations is predominantly influenced by air masses coming from the ocean (Fig. S1), supposedly cleaner
than the air masses coming from the continent with anthropized areas.

In SaoPaulo
˜ and Belo Horizonte, the concentrations of toluene, ethylbenzene, formaldehyde and acetaldehyde stood out, with higher values between August/2017, November/
2017 and June/2018, probably resulting from the unfavorable conditions for dispersion of pollutants in this period (da Rocha et al., 2023). Associated with this, Sao ˜ Paulo state,
which is the largest sugarcane producer in the country, is influenced by the burning of this raw material in the harvestand post-harvest periods (burning of bagasse and land for
replanting), emitting several pollutants/ contaminants such as NOx and VOCs (Hall et al., 2012; Francisco et al., 2016). In the C3 campaign, Sao ˜ Paulo presented a significant
increase in the concentration of acetaldehyde, generated in the atmosphere by the oxidation of ethanol. By analyzing the reverse trajectories of air masses (Fig. 3(e)), it was
possible to perceive that, during this period, air masses that passed through Ribeirao ˜ Preto and Piracicaba, cities with a large production of sugarcane and derivatives, arrived
at the sample site and, as a function of the bagasse burning period, it certainly justifies the high levels of acetaldehyde, even in a period of greater solar irradiation, which
facilitates the photochemical removal of VOCs (Andreae and Merlet, 2001; Hall et al ., 2012; Francisco et al., 2016).

Despite being the city with more defined seasons among those studied, Londrina was the one that presented the lowest seasonal variation; This can be explained
by the unusually high rainfall throughout the year, which mischaracterizes seasonality, reflecting little variation in concentration levels.

Compared to the levels reported from other cities around the globe, the atmospheric levels of BTEX, formaldehyde and acetal-dehyde obtained in this study are
within the same order of magnitude (Alfoldy et al., 2019; Salameh et al., 2019; Yang et al., 2019; Cruz et al., 2020).

3.4. Atmospheric concentration of other inorganic (HCl and H2S) and organic (HCOOH and CH3COOH) acids

Mean concentrations ranged from <0.93–15, <0.24–3.8, <0.54–5.9, 2.4–9.7 ÿg mÿ 3 for HCl, H2S, HCOOH and CH3COOH, respectively. Fig. 3(f) shows the

variation in the concentrations of these pollutants over the sampling period. Table S4 presents the average concentrations obtained for these pollutants with the respective quantification limits of the passive samplers.

Salvador, S˜ ao Paulo and Londrina had HCl concentrations above the recommended by the Texas Quality Criterion for ambient air

8
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

3
(TCEQ, 2016), of 7.9 ÿg mÿ , in at least one of the sampling periods; however, in no case, it was above the Ontario Ministry of the (OME, 2020).
3
Environment (OME) recommendation of 20 ÿg mÿ
From Fig. 3(f), it is observed that, after S˜ ao Paulo and Londrina which, despite not being coastal cities, had air masses arriving at their sampling points (Figs. 4
and S1) with marine influence in some periods, the highest concentration of HCl occurred in Salvador in the dry period. In this location, the source of natural HCl
emission should explain these concentrations since, in all campaigns, this location received air masses from the ocean, in addition to known anthropogenic sources of
this acid not having been identified there.
Urban areas that are influenced by marine aerosol are more susceptible to interactions of acids such as H2SO4 and HNO3 with NaCl from marine spray, increasing
the concentration of HCl(g) in the atmosphere, according to Eq. (1). ˜ Paulo, with values below the limit recommended by
For H2S, the highest concentrations were in Belo Horizonte and Sao Quality Criteria
the Texas ) and the
3 3
for ambient air (7 ÿg mÿ range recognition of this ) (TCEQ, 2016), but higher than the level of odor perception (0.2 to 2 ÿg mÿ
3
odorous gas (0.6–6.0 ÿg mÿ higher than the measured ) (CEN, 2022). Furthermore, these values found in urban areas are 3 to 6 times
concentrations in the Buracica-BA oil field in 2005 ( Campos et al., 2010) using the same methodology, which warns about the significant presence of this gas in those
two urban environments.
Hydrogen sulfide (H2S) originates in urban areas mainly due to poor sanitation and the frequent presence of open sewers, where the microbiological decomposition
of organic matter produces this gas with a characteristic unpleasant odor. In the samples, this air contaminant was only found at a significant concentration and
frequency in 3 of the 5 urban centers studied: Belo Horizonte, Sao ˜ Paulo and Rio de Janeiro. Among them, Belo Horizonte (BH) presented the highest concentrations,
possibly due to the proximity of the sampling site to Lake Pampulha, currently polluted by the discharge of untreated sewage (WHO, 2000; WHO, 2003) and brought
by the Sarandi and Ressaca streams, with springs in Contagem, Metropolitan Region of BH, accounting for about 70% of this pollution (IMGA, 2018). Furthermore, as
already discussed, the site is close to 2 landfills that should also contribute to the increase in H2S levels in that atmosphere. In Sao ˜ Paulo, the most expressive levels
of H2S concentration are clearly explained by the proximity of the sampling point to the banks of the Pinheiros River which, according to the Report on Inland Waters
of the State of Sao ˜ Paulo, by CETESB in 2018 (CETESB, 2019), presented a Quality Index of Water classified from bad to very bad due to the presence of high
organic load caused by domestic and industrial sewage dumps over the years. In Rio de Janeiro, although the sample site is not close to open sewage disposal sites
and/or sanitary landfills, the H2S concentrations measured can be possibly justified by the presence of an Experimental Sewage Treatment Center at the Federal
University do Rio de Janeiro (UFRJ ), located next to the sampling site (~ 300 m). Salvador and Londrina basically presented values below the quantification limit of
the sampler.

Organic acid concentrations were also higher in Sao ˜ Paulo and Belo Horizonte. However, for these compounds, recommendations/legislation have not
yet been established, so that the evaluation of these levels must be based on the values obtained in other studies on the behavior of these acids in an urban
environment.

Fig. 4. Reverse trajectories of air masses coming from the ocean that landed in urban centers: (a) Sao ˜ Paulo, (b) Londrina.

9
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Fig. 3(f) shows that the concentrations determined for formic acid and acetic acid showed little variation due to seasonality, when compared to the
values found for hydrochloric and nitric acids, but with a predominance of increased concentrations in the dry period for all urban centers. Acetic acid
was the most abundant species.
Sao
˜ Paulo was the urban center with the highest levels of formic and acetic acids, followed by Belo Horizonte. In general, the highest concentrations
of formic acid were observed in periods of more intense solar irradiation, which was expected, due to the favoring of the photochemical formation of
this compound from the ozonolysis of olefins and/or formaldehyde in the presence of free radicals. These higher levels partially reflect a contribution
from in situ formation.
On the other hand, the burning of fuels in vehicle engines is also a source of carboxylic acids in the atmosphere at levels considered important in
urban environments. These emissions release different organic acids into the atmosphere, of which 78% correspond to acetic and formic acids
(Grosjean et al., 1990; Cruz et al., 2019). The ratio between the concentrations of these acids (FA/AA) suggests their atmospheric origin, evidencing
the predominance in the formation of carboxylic acids in situ through photochemical reactions for ratios >1 and direct emissions for ratios smaller than
1. The ratios for these acids are shown in Fig. 5.
It is observed that 71% of the data presented ratios <1, indicating direct emission as the main source of organic acids in these urban centers, as
a function of emissions. Sampling campaigns C3 and C5 in Salvador, C4 and C5 in Sao ˜ Paulo, C4 in Belo Horizonte and C6 in Rio de Janeiro showed
ratios >1 and were favored by low rainfall intensity and intense solar irradiation, generating an increase in these acids in the atmosphere by
photochemical formation, prevailing in situ formation.
Fig. 6 shows the correlation between the concentrations of the two organic acids obtained in the urban centers considered in this study. As
expected, the correlation coefficient indicates that the source of these acids in the atmosphere is predominantly the same (Cruz et al., 2019),
suggesting vehicle emissions as the main source. However, there may be contributions from direct natural emissions and photochemical reactions in
the atmosphere influencing these concentrations on a smaller scale.
The concentrations of organic acids found in this study (Table S4) are in the same order of magnitude as other studies: in S˜ ao Paulo (Montero
3 3
(Brazil), 1.1–36 ÿg mÿ and 0.20–26 ÿg mÿ et al., 2001); 0.64–12 ppbv and 0.51–11 ppbv (Souza and Carvalho, 2001); in
3 3
Southern Region (Vietnam), 5.4–16 ÿg mÿ and 5.5–21 ÿg mÿ (Lan et al., 2004); in Colorado (USA), 1.0–3.6 ppbv (Mattila et al., (Cruz et
3 3
2018); in Salvador (Brazil), 0.90–6.2 ÿg mÿ exception and 0.48–8.6 ÿg mÿ al., 2019) for HCOOH and CH3COOH, respectively. With the
of the most recent measurements made by Mattila et al. (2018) and Cruz et al. (2019), little is currently studied about organic acids in the gaseous phase in the
atmosphere, although they participate in important reactions in the atmosphere, in addition to their contribution to acid rain.

Regarding the individual contribution of acids measured in urban centers whose atmosphere was studied, acetic acid was the most abundant
species in most urban centers with the exception of Salvador, contributing with approximately 32% in relation to the other acids studied. This higher
percentage of acetic acid, even compared to formic acid, corroborates the statement that vehicle emissions were more influential than the
photochemical formation of these compounds in the atmosphere.
In Salvador, HCl was the most abundant acid (36%), probably due to the reaction between gaseous HNO3 and marine spray
particles, a process previously mentioned.
Considering the nature of the acids, it is observed that the atmosphere of the studied urban centers did not present a significant difference
between the percentages of organic and inorganic acids.

3.5. Linear correlation matrix

Pearson correlation for atmospheric pollutants/contaminants was carried out in order to homologate/assist the identification of the most significant
interactions and emission sources and to confirm the statements previously made throughout this study and with the relationships proposed in the
multivariate analysis that is part of this study. Pearson correlation matrices between the studied pol-lutants and by urban centers are presented in
Tables S5 to S9. It can be observed that NO2 and NO did not correlate with each other. This indicates equilibrium between the oxidation of NO to
NO2 and the rate of NO2 photosynthesis, resulting in a stationary process dependent on meteorological conditions.

In Salvador (Table S5), S˜ ao Paulo (Table S6) and Rio de Janeiro (Table S9), nitrogen oxides show a strongly positive correlation

Fig. 5. Ratio between formic and acetic acid (FA/AA) concentrations in Brazilian urban centers (2017–2018).

10
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Fig. 6. Correlations between HCOOH and CH3COOH concentrations in the five Brazilian urban centers (2017–2018).

with NH3, highlighting the role of these compounds in the formation cycle of NH2*, suggesting that ammonia acted as a precursor of nitrogen oxides. In Londrina
(Table S8) and Rio de Janeiro (Table S9), strong negative correlations between NO2 and O3 were highlighted, indicating tropospheric ozone removal and HNO3
formation, which is reinforced by the positive correlation between NO2 and HNO3 in those two urban centers. In Sao ˜ Paulo and Londrina, positive correlations
between NO2 and acids (H2S and HNO3) were also highlighted, which may indicate a similarity in the emission sources of these compounds.

Strong correlations between SO2 and some of the acids were also observed at all sites. The high correlation between these com-pounds suggests similar
atmospheric behavior and possibly the same source, especially considering cities such as Salvador and Rio de Janeiro, very prone to photochemical reactions and
the presence of marine aerosol, where SO2 correlates with HNO3. SO2 also showed a positive correlation with O3 in S˜ ao Paulo and Rio de Janeiro; this must
indicate that these pollutants and/or their precursors have their main origin in vehicle emissions in these locations.

In Salvador's atmosphere, O3 showed negative correlations with NO, HCOH and xylenes. With NO, the negative correlation can be justified by the action of O3
as the predominant oxidant in the conversion of NO into NO2 (Eq. (2)). HCOH and xylenes should correlate with O3 due to the removal of these VOCs by
photochemical reaction.

NO + O3ÿNO2 + O2 (2)

According to the reactivity data evaluated on the MIR (Maximum Incremental Reactivity) scale (Carter, 2010), Salvador was the urban center with the lowest
tropospheric ozone formation potential among the locations considered in this study, which justifies the removal of VOCs and NO from the atmosphere as the most
significant process. In the case of VOCs, this occurs mainly by photochemical reactions (Atkinson, 2000) and, for NO, by the consumption of ozone in the conversion
of NO to NO2 (Eq. (2)) (Fowler et al., 2008).
On the other hand, the correlations of O3 with ethylbenzene in Rio de Janeiro, and HCOH and xylenes in Belo Horizonte (Table S7), were positive, demonstrating
that, in these cases, the reactivity of these compounds in the atmosphere may have favored the production of tropospheric ozone.

In Rio de Janeiro, O3 showed a strong negative correlation with NH3, indicating gas-particle transformation, consuming O3 in
ammonium nitrate formation reactions in the troposphere (Eqs. (3) and (4)).

NH3 + HO* ÿ H2N* + H2O (3)

H2N* + O3 ÿ NOx + H2O (4)

Among other more abundant nitrogen compounds in the atmosphere, ammonia stands out as the most important alkaline gaseous species, acting as a key
species in neutralizing the acidity of nitrate and sulfate aerosols (Eq. (5)) (Felix and Cardoso, 2004). This was shown in Sao ˜ Paulo, the place with the highest
concentration HNO3,
of with a strong negative correlation with NH3 (Table S6).

NH3 + HNO3 ÿ NH4NO3(part.) (5)

Salvador and Rio de Janeiro were the only cities to present strongly positive correlations between HNO3 and HCl, which was expected due to the proximity of
these urban centers to the sea, receiving more intense contributions from marine spray, reacting with nitric acid, generating particulate matter and gaseous HCl.

these˜ Paulo and Belo Horizonte, there was a positive correlation between toluene and H2S, certainly explained by the proximity of In Sao emission sources, characteristic of
compounds, such as the discharge of domestic and industrial sewage into the Pinheiros river – SP and Lake Pampulha – MG, related to H2S and emissions from landfills linked
to toluene (WHO, 2000; WHO, 2003; Mrowiec, 2014; Hajizadeh et al., 2018). In Londrina, H2S correlated directly with NH3, which may indicate a predominance of direct
emissions of these pollutants by decomposition of organic matter (NPI, 2014; Shi et al., 2015); the opposite was observed in Rio de Janeiro, where the burning of fossil fuels and/
or industrial processes involving NH3 seemed to play a more important role, resulting in a negative correlation with H2S. Furthermore, in Belo Horizonte, H2S was positively
correlated with formic acid, indicating significant biogenic emissions.

VOCs were frequently observed to present an important negative correlation with nitrogen oxides, stronger in S˜ ao Paulo, Rio de

11
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Janeiro and Belo Horizonte, which may be associated with the consumption of these compounds in the formation of ozone or other
products involved in photochemical smog, typical of polluted urban atmospheres. In Belo Horizonte and Rio de Janeiro, aldehydes
show a strong positive correlation with their respective organic acids (formic and acetic), indicating important secondary formation. In
Salvador, formaldehyde and toluene showed a positive correlation, as well as o-xylene and m,p-xylene with each other, indicating
that these pollutants can be emitted by the same source, probably vehicles. The direct correlation found between aldehydes in Rio
de Janeiro also contributes to the assertion that these compounds have predominantly been emitted by the same source (vehicle).
Among BTEX, xylenes were the VOCs that most correlated with other pollutants, probably due to the high rate of their reactions in
the troposphere.
Londrina was the city with the highest number of significantly positive correlations between the studied pollutants, mainly regarding
those characteristic of vehicle emissions in urban areas such as NO2, SO2 and aldehydes, showing the little variation of emission
sources in this place, when compared to other urban centers.

3.6. Multivariate analysis

In the Principal Component Analysis (PCA), score plots (cases) and loading plots (variables) were obtained from the correlations
between the selected variables. Data were pre-processed by autoscaling, eliminating the influence of the difference in orders of
magnitude between them.
In the analysis, 5 locations were considered (SA – Salvador; SP – S˜ ao Paulo; LD – Londrina; RJ – Rio de Janeiro; BH – Belo
Horizonte), 6 sample periods (from 08/2017 to 06/2018), concentrations of 16 pollutants/contaminants and meteorological data
(rainfall, relative humidity, solar radiation, temperature, wind speed and direction). Londrina did not have INMET's automatic network
of meteorological measurements and it was not possible to have data on solar irradiation. Therefore, two data matrices were
generated: one considering Londrina and the other without Londrina, but with solar irradiation data for all other locations and periods.
After applying PCA to the explored data, it was decided how many principal components (PCs) would be used in the evaluation of

Fig. 7. Score plots: (a) with the variable Londrina; (b) without the variable Londrina associated with loading plots: (c) with the variable Londrina (d) without the
variable Londrina.

12
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

the similarity between the objects (samples). According to the screen-plot criterion, after a certain number of PCs, data variance re-flects only the residuals of their
random fluctuations, graphically identified by the inflection. Fig. S2(a) and (b) show the screen-plots for the data under analysis. It is observed that the inflection occurs
in the second PC for both plots, indicating that the first two PCs can be considered sufficient in assessing the similarity between the samples. Considering Londrina,
PC1 and PC2 can account for approximately 43% of data variability. Without this variable (Londrina) and with the inclusion of the solar irradiation variable, PC1 and
PC2 account for 45%.

Fig. 7 presents the score plots (a and b), which represent the linear projection of the objects (in this case, the locations where the data were collected) on the plan
formed by the first two PCs, as well as the loading scores (c and d) that provide the projection of the variables on the first two PCs.

˜
In Fig. 7(a), it is possible to observe a tendency for the samples to form three groups. Two groups formed by two locations (Sao Paulo and Belo Horizonte) and
one group formed by three locations (Salvador, Londrina and Rio de Janeiro). In the latter, there is a tendency to separate Rio de Janeiro. Salvador and Londrina
showed a lot of similarity between the samples, probably due to the fact that Londrina is the smallest urban center, presenting less diversity of emission sources; it is
more similar to the group formed by Rio de Janeiro and Salvador, which presented lower concentration levels.

The same two groups (S˜ ao Paulo and Belo Horizonte) were formed in Fig. 7(b); in this case, without the presence of Londrina.
Collection date did not influence group separation. This separation demonstrates the direct relationship of similarity with the typical sources of each group and
meteorological influences. S˜ ao Paulo and Belo Horizonte have a very similar climate, with a dry winter and a rainy summer; they are urban centers with intense
vehicle and industrial emissions, positioned close to specific sources such as open sewage discharges and sanitary landfills, which contribute to these two centers
generally presenting greater urban atmospheric pollution.

Rio de Janeiro and Salvador presented similar climatological conditions throughout the sampling periods and, in addition, they are the two cities closest to the sea,
consequently with a greater influence of marine aerosol, with potentially cleaner air masses arriving at the sampling points and with better conditions for dispersion of
pollutants, when compared to S˜ ao Paulo and Belo Horizonte.
As already discussed, the reverse trajectories of air masses showed that the direction of the winds in the sampling periods greatly influenced atmospheric
concentrations. Rainfall contributed little to data modeling, corroborating the previous statements, in the evaluation of the concentrations as a function of seasonality,
where it was observed that solar irradiation and temperature were much more significant to explain the behavior of these pollutants/contaminants in the troposphere
than rainfall.
The set of graphs (Fig. 7) highlights that the Sao that is, ˜it Paulo samples showed that this was the point most correlated with most pollutants,
was the place with the highest concentration levels.
The variables responsible for separating the group were HCl, NO, SO2, benzene, O3, CH3COH, NOx, NO2, HCOOH and CH3COOH, compounds predominantly
emitted by vehicle emissions in urban areas. That is, the highest concentration levels observed for these compounds were directly influenced by fuel burning.

Temperature and humidity were responsible for the separation of the group involving Salvador, Londrina and Rio de Janeiro in Fig. 7(c), indicating that, for these
urban centers, in situ secondary formation also contributed to the levels of pollutants found, although vehicles emissions contribute more substantially.

Fig. 7(d) shows that solar irradiation was important for differentiating the groups represented by Salvador and Rio de Janeiro, indicating that, in Salvador,
photochemical reactions strongly influenced the composition of the atmosphere, also observed in the seasonal variation of pollutants.

In general, the multivariate analysis shows that VOCs and organic acids are more influenced by the parameters temperature and solar irradiation which, in turn, is
in the opposite quadrant to nitrogen oxides, showing less influence in this case. The strongest direct relationship of NOx was with wind direction, showing competition
between local emission and photochemical aging of these compounds brought from other locations.

4. Conclusions

In general, considering all the contaminants/pollutants measured in the five urban centers, Sao ˜ Paulo (SP) presented higher
concentrations (O3, SO2, NO2, NOx, HNO3, HCl, Benzene, Ethylbenzene, oxylene, CH3COH, HCOOH and CH3COOH) in 76% of the cases. This was expected, as
this urban center has the largest vehicle fleet in the country and is more industrialized. Belo Horizonte (BH) showed higher concentrations in 23% of cases, which
include CH3COH, H2S, NH3 and Toluene. The latter stood out in Belo Horizonte, where it showed an average concentration of 3 to 19 times higher than in SP and
RJ, respectively. There are emissions from landfills in BH at two points close to the sampling site, where toluene is emitted by burning biogas, in addition to other
emissions from solvents, paints and foams present in the polluted Lake Pampulha waters, which must be driven to the sampling site by prevailing winds.

Londrina had a higher concentration of O3, SO2, ethylbenzene and HCl than other locations, with the exception of SP, even with the smallest vehicle fleet among
the studied urban centers. The Londrina sampling site has an intense flow of heavy vehicles, besides the contribution of air masses coming predominantly from other
urban areas.
Salvador and Rio de Janeiro did not show atmospheric contaminants with significantly higher concentrations than the other lo-cations. The first was the city with
the lowest average concentrations. It has the second smallest vehicle fleet and the atmosphere is constantly influenced by potentially cleaner air masses coming from
the Atlantic. This justifies its similarity with RJ, both coastal cities with better conditions for the dispersion of pollutants. The concentrations of HNO3 in the atmosphere
of these two urban centers were the lowest, indicating that this acid reacts favorably with NaCl from marine spray particles, producing fine particulate matter and

13
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

gaseous HCl.
In relation to legislated levels or recommendations, the concentration of SO2 in the atmosphere of SP, Londrina and RJ exceeded the
limit of 40 ÿg mÿ 3 recommended in Brazil in three sampling periods in SP and eleven in Londrina and Rio de Janeiro. Although the O3
3 3
nationally legislated limit for O3 (140 ÿg mÿ one ) was not exceeded, Sao ˜ Paulo presented a concentration of 139 ÿg mÿ integrated in
of the sampling periods for 7 days.
Formaldehyde concentrations exceeded the most restrictive value legislated by the Texas Commission on Environmental Quality (ÿ
3.3 ÿg mÿ 3) in most cases. Benzene showed mean concentrations higher than 1.7 ÿg mÿ associated 3
, by WHO with the probability of
3
leukemia (1:100,000) in all locations; in S˜ ao Paulo, the annual European Union standard of 5.0 ÿg mÿ ˜ Paulo was also exceeded.
Salvador, Sao and Londrina had HCl concentrations above the limit recommended by the current Texas Air Quality Criterion in at least one of
3 3 exceeded.
of 7.9 ÿg mÿ the sampling periods, but in none of the cases was the OME recommendation of 20 ÿg mÿ ˜ Paulo: H2S was found
The highest concentrations of H2S and NH3 occurred in Belo Horizonte and Sao above the odor perception ), demonstrating that the
3 3
level (0.2 to 2 ÿg mÿ and ) and recognition range (0.6–6.0 ÿg mÿ pollution of the Pinheiros River in S˜ ao Paulo
from Lake Pampulha in Belo Horizonte may be impacting air quality from the nuisance of the perception of bad smell. Since the Pinheiros River also receives industrial
sewage discharges, it also contributes to the increase in toluene concentrations in the air, according to water quality reports in the State of Sao ˜ Paulo.

Formic acid/acetic acid ratios were found <1 in 71% of cases, indicating that organic acids are primarily emitted directly into the atmosphere
in these urban centers, probably due to vehicle emissions, showing less important photochemical formation. Among the acidic species
determined, acetic acid was the most abundant in most urban centers. However, the contribution of organic and inorganic acids to atmospheric
acidity was similar for all sites.
Strong positive correlations were observed among nitrogen pollutants, conventional pollutants (SO2, NO2 and O3) and VOCs among
themselves, indicating a significant influence of vehicle emissions for the observed concentration levels.
The loading charts identified rainfall and HNO3 as the variables with the least contribution to data modeling, corroborating what was found in
the seasonal assessment of concentrations, where solar irradiation and temperature were indicated as more influential conditions than rainfall,
especially in Salvador and Rio de Janeiro . ˜ Paulo showed greater
Data from Saoexpansion in the graph with pollutants/pollutants predominantly emitted by vehicle emissions in urban areas, suggesting that
the highest levels observed for these compounds are directly influenced by the burning of vehicle fuels. The variables toluene, HCOH, o-xylene,
m,p-xylene, H2S and wind speed separated Belo Horizonte by inferring the significance of emissions caused by volatilization of pollutants from
open sewers and industrial waste.
VOCs and organic acids were more influenced by temperature and radiation, and nitrogen oxides by wind direction; this indicates the
contribution of photochemical reactions for VOCs and organic acids and the relationship between local emission and photo-chemical aging for
nitrogen oxides.
The concentration levels of formaldehyde and benzene, which exceed international recommendations in the atmosphere of Brazilian urban centers, may alert
to the need to include national limits for these compounds, since they are pollutants with well-established risks.

The use of passive sampling, using AnalyzeAr kits, showed adequate performance for measurements of 16 pollutants/contaminants
simultaneously in five urban centers, confirming the possibility of its use for monitoring air quality at a low cost, thus facilitating compliance with
national control plans with regard to air quality.
Future studies using the same low-cost methodology are recommended, expanding the measurement scale both temporally and spatially,
characterizing air quality in strategic areas, associating the main pollutants with population's health. These studies will also contribute globally
by improving the coverage of global monitoring of air pollutants, consequently increasing the accuracy of estimating exposure in the various
places where people live.

CREdiT authorship contribution statement

All authors contributed to the study conception and design. Franciele O. Campos da Rocha: Writing - original draft, Concep-tualization,
ˆ
Data curation, Formal analysis, Investigation, Methodology, Validation. Vania P. Campos: Supervision, Writing - review & editing, Funding
acquisition. Gisele O. da Rocha: Supervision, Writing - review & editing. Marcos A. Bezerra: Data curation and Writing - review & editing.
Anne Valesca S. Brito: Data curation, Methodology. Indiara dos Santos Sampaio: Data curation, Methodology. All authors reviewed the
manuscript.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Data availability

Data will be made available on request.

14
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Acknowledgements
´
The authors would like to thank the Conselho Nacional de Desenvolvimento Científico e Tecnologico (CNPq), Fundaç˜ ao de Amparo Pesquisa do Estado da
`
to Bahia (FAPESB), the National Institute of Science and Technology (INCT) for Energy & Environment (INCT E&A, project CNPq 465497/2014-4), the
Interdisciplinary Center for Energy and Environment (CIEnAm) from Universidade Federal da Bahia and Pro-Pesquisa/PRPPG-Universidade Federal de Roraima.
´
The authors are also grateful for Professors P´erola Vasconcellos (USP-SP), Zenilda Cardeal (UFMG-MG), Graciella Arbilla (UFRJ-RJ) and Cristina Solci (UEL-PR),
for the collaboration on sampling campaigns. FOCR thanks FAPESB for her scholarship (nÿ BOL0741/2016). AVSB and ISS would like to thank PIBIC and
PERMANECER/UFBA for their scholarships. GOR is grateful for their research fellowship from CNPq.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.uclim.2023.101739.

References

Abe, K., Miraglia, S., 2016. Health impact assessment of air pollution in S˜ ao Paulo, Brazil. Int. J. Environ. Res. Public Health 13, 694. https://doi.org/10.3390/
ijerph13070694.
Agudelo-Castaneda, ˜ DM, Teixeira, EC, Rolim, SBA, Pereira, FN, Wiegand, F., 2013. Measurement of particle number and related pollutant concentrations in an urban area in South
Brazil. Atmos. Environ. 70, 254–262. https://doi.org/10.1016/j.atmosenv.2013.01.029.
Alfoldy, B., Mahfouz, MMK, Yigiterhan, O., Safi, MA, Elnaiem, AE, Giamberini, S., 2019. BTEX, nitrogen oxides, ammonia and ozone concentrations at traffic influenced and background
urban sites in an arid environment. Atmos. Pollut. Res. 10, 445–454. https://doi.org/10.1016/j.apr.2018.08.009.
Andreae, MO, Merlet, P., 2001. Emission of trace gases and aerosols from biomass burning. Glob. Biogeochem. Cycles 15 (4), 955–966. https://gfmc.online/vfe/
Andreae-and-Merlet%202001.pdf (accessed August 25, 2023).
Andre˜ ao, WL, Albuquerque, TTA, Kumar, P., 2018. Excess deaths associated with fine particulate matter in Brazilian cities. Atmos. Environ. 194, 71–81. https://
doi.org/10.1016/j.atmosenv.2018.09.034.
´
Anglada, JM, Martins-Costa, MT, Francisco, JS, Ruiz-Lopez, MF, 2020. Reactivity of undissociated molecular nitric acid at the air–water interface. J. Am. Chem.
Soc. 143 (1), 453–462. https://doi.org/10.1021/jacs.0c11841.
Atkinson, R., 1994. Gas-Phase Tropospheric Chemistry of Organic Compounds, Second ed.
Atkinson, R., 2000. Atmospheric chemistry of VOCs and NO(x). Atmos. Environ. 34, 2063–2101. https://doi.org/10.1016/S1352-2310(99)00460-4.
Barbosa, JL, Campos, VP, Campos da Rocha, FO, 2023. Development of a passive sampler for nitrogen oxides (NOx) in the atmosphere. Rev. Virtual Quim. 15,
332–333. https://doi.org/10.21577/1984-6835.20220132.
Behera, SN, Mukesh, S., Mishra, PK, Nayak, P., Damez-Fontaine, B., Tahon, R., 2015. Passive measurement of NO2 and application of GIS to generate spatially-
distributed air monitoring network in urban environment. Urban Climate. 14, 396–413. https://doi.org/10.1016/j.uclim.2014.12.003.
Brazil, 2018. Resolution No. 491/2018. Air Quality Standards - National Environmental Council (CONAMA). https://pesquisa.in.gov.br/imprensa/jsp/visualiza/
index.jsp?data=11/21/2018&journal=515&page=155 (accessed March 23, 2023).
Brazil, 2019a. ANP - Statistical Yearbook 2018 - National Petroleum Agency. https://www.gov.br/anp/pt-br/centrais-de-conteudo/publicacoes/anuario-estatistico
(accessed February, 02, 2023).
Brazil, 2019b. Brazilian Agriculture and Livestock in Numbers - Ministry of Agriculture, Livestock and Supply. https://www.gov.br/agricultura/pt-br/assuntos/
agricultural-politics/all-publicacoes-de-agricultural-politics/agropecuaria-brasileira-em-numeros (accessed March 23, ˜2023).
Campos, VP, Cruz, LPS, Alves, ˜EM, de Santos, TS, Silva, AD, dos Santos, ACC, Lima, AMV, Paixao, CS, Santos, DCMB, Brandao, ˜DS, Andrade, EJS, KCS, de Ramos, MS, Pontes,
influenced areas as a practice MCG, do Amaral, MF, Mattos, RR, 2006. Passive atmospheric monitoring of SO2, NO2 and O3 in Moreira Jr., JI, Conceiçao, urban and industrially
´ ´
of environmental chemistry for graduating students. ˜ Quim Nova 29, 872–875. https://doi.org/10.1590/S0100-
40422006000400042.
Campos, VP, Cruz, LPS, Godoi, RHM, Godoi, AFL, Tavares, TM, 2010. Development and validation of passive samplers for atmospheric monitoring of SO2,
NO2, O3 and H2S in tropical areas. Microchem. J. 96, 132–138. https://doi.org/10.1016/j.microc.2010.02.015.
Carter, WPL, 2010. Updated Maximum Incremental Reactivity Scale and Hydrocarbon Bin Reactivities for Regulatory Applications. https://www.arb.ca.gov/regact/
2009/mir2009/mir10.pdf (accessed August 25, 2023).
CEN - European Committee for Standardization, 2022. Stationary Source Emissions - Determination of Odor Concentration by Dynamic Olfactometry and Odor Emission Rate. https://
www.en-standard.eu/din-en-13725-stationary-source-emissions-determination-of-odour-concentration-by-dynamic-olfactometry-and- odor-emission-rate/ (accessed March 21,
2023).
CETESB - Companhia de Tecnologia e Saneamento Ambiental de S˜ ao Paulo, 2019. State of Sao ˜ Paulo ´ Inland Water Quality Report - 2018. https://cetesb.sp.gov.br/aguas-interiores/wp-
´
content/uploads/ sites/12/2019/10/Relatorio-de-Qualidade-das- Aguas-Interiores-no-Estado-de-SP-2018.pdf (accessed August 25, 2023).
Conforti, A., Mascia, M., Cioffi, G., de Angelis, C., Coppola, G., de Rosa, P., Pivonello, R., Alviggi, C., de Placido, G., 2018. Air pollution and female fertility: a
systematic review of literature. Play Biol. Endocrinol. 16, 117. https://doi.org/10.1186/s12958-018-0433-z.
Correa, S., Arbilla, G., Monteiro, E., Quit´erio, S., de Souza, C., Gatti, L., 2010. Five years of formaldehyde and acetaldehyde monitoring in the Rio de Janeiro
downtown area – Brazil. Atmos. Environ. 44, 2302–2308. https://doi.org/10.1016/j.atmosenv.2010.03.043.
Cruz, L., Mota, E., Campos, V., Santana, F., Luz, S., Santos, D., 2019. Inorganic and organic acids in the atmosphere of the urban area of the city of Salvador, Brazil .
J. Braz. Chem Soc 30 (1), 904–914. https://doi.org/10.21577/0103-5053.20180227.
Cruz, LPS, Santos, DF, dos Santos, IF, Gomes, IVS, Santos, AVS, Souza, KSPP, 2020. Exploratory analysis of the atmospheric levels of BTEX, criteria air pollutants and meteorological parameters in a tropical
urban area in northeastern Brazil . Microchem. J. 152, 104265–104274. https://doi.org/10.1016/j. microc.2019.104265.

Cruz, LP, Alves, RS, da Rocha, FO, Moreira, MS, dos Santos Júnior, A., 2022. Atmospheric levels, multivariate statistical study, and health risk assessment of odorous compounds (H2S
and NH3) in areas near polluted urban rivers in the city of Salvador, in northeastern Brazil. Air Qual. Atmos. Health 15, 159–176. https://doi.org/10.1007/s11869-021-01095-7.

Cyrys, J., Eeftens, M., Heinrich, J., Ampe, C., Armengaud, A., Beelen, R., Bellander, T., Beregszaszi, T., Birk, M., Cesaroni, G., Cirach, M., de Hoogh, K., De Nazelle, A., de Vocht, F.,
Declercq, C., Dedele, A., Dimakopoulou,
¨ K., Eriksen, K., Galassi, C., Grauleviÿciene, R., Grivas, G., Gruzieva, O., Gustafsson, A.H., Hoffmann, B., Iakovides, M., Ineichen, A.,
Kramer, U., Lanki, T., Lozano, P., Madsen, C. ., Meliefste, K., Modig, L., Molter, ¨ A., Mosler, G., Nieuwenhuijsen, M., Nonnemacher, M., Oldenwening, M., Peters, A., Pontet, S.,
Probst-Hensch,
´ N., Quass, U., Raaschou-Nielsen, O., Ranzi, A. ., Sugiri, D., Stephanou, EG, Taimisto, P., Tsai, MY, Vaskovi, ¨ E., Villani, S., Wang, M., Brunekreef, B., Hoek,
G., 2012. Variation of NO2 and NOx concentrations between and within 36 European study areas: results from the ESCAPE study. Atmos. Environ. 62, 374–390. https://doi.org/
10.1016/j.atmosenv.2012.07.080.

15
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

da Rocha, FOC, Barbosa, JL, Campos, VP, Alves, RS, 2022. Nitrogen oxides levels in the atmosphere of different Brazilian urban centers, by passive sampling.
J. Braz. Chem Soc 33, 143–156. https://doi.org/10.21577/0103-5053.20210131.
da Rocha, FOC, Campos, VP, da Rocha, GO, Brito, AVS, Sampaio, IS, 2023. Characterization of airborne gaseous BTEX and aldehydes from populated Brazilian cities as representative atmospheres of typical urban
areas from the southern hemisphere. Air Qual. Atmos. Health 1, 1–19. https://doi.org/10.1007/s11869-023- 01341-0.

´
Dancey, C., Reidy, J., 2006. Semi-Mathematical Statistics for Psychology: Using SPSS for Windows, Third ed. Artmed, Porto Alegre.
De S´ a Borba, PF, Martins, EM, Ritter, E., Corrˆea, SM, 2017. BTEX emissions from the largest landfill in operation in Rio de Janeiro, Brazil. Bull. Environ. Contam.
Toxicol. 98, 624–631. https://doi.org/10.1007/s00128-017-2050-5.
Di Carlo, P., Kajii, Y., Pitari, G., Tuccella, P., Dari-Salisburgo, C., Busilacchio, M., Aruffo, E., Biancofiore, F., 2014. Wildfires impact on surface nitrogen oxides and
ozone in Central Italy. Atmos. Pollut. Res. 6, 29–35. https://doi.org/10.5094/apr.2015.004.
European Union - Directive, 2008. Directive 2008/50/EC of on Ambient Air Quality and Cleaner Air for Europe. https://eur-lex.europa.eu/legal-content/EN/TXT/
PDF/?uri=CELEX:32008L0050&from=en (accessed August 25, 2023).
Fang, J.-J., Yang, N., Cen, D.-Y., Shao, L.-M., He, P.-J., 2012. Odor compounds from different sources of landfill: characterization and source identification. Waste Manag. 32,
1401–1410. https://doi.org/10.1016/j.wasman.2012.02.013.
Fantozzi, F., Monaci, F., Blanusa, T., Bargagli, R., 2015. Spatio-temporal variations of ozone and nitrogen dioxide concentrations under urban trees and in a nearby
open area. Urban Climate. 12, 119–127. ˆ https://doi.org/10.1016/j.uclim.2015.02.001. ´
Felix, EP, Cardoso, AA, 2004. Atmospheric ammonia (NH3) : Fontes, transformaç˜ ao, sorvedouros em´etodos de analise. Quim Nova 27, 123–130. https://doi.org/ 10.1590/
S0100-40422004000100022.
Fowler, D., Amann, M., Anderson, R., Ashmore, M.R., Depledge, M., Derwent, D., Grennfelt, P., Hewitt, C.N., Hov, O., Jenkin, M., Kelly, FJ, Liss, P., Pilling, MJ, Pyle, J., Slingo, J.,
Stevenson, D., 2008. Ground-Level Ozone in the 21st Century: Future Trends, Impacts and Policy Implications Science Policy. https://royalsociety.org/~/media/
royal_society_content/policy/publications/2008/7925.pdf (accessed August 25,ˆ 2023). ˆ
Francisco, AP, Alvim, D., Gatti, L., Pesquero, C., Assunç˜ ao, JV, 2016. Tropospheric ozonium and organic composts vol´ ateis em regi˜ ao impacted by agroindustry
canavieira. Quim Nova 39, 1177–1183. https://doi.org/10.21577/0100-4042.20160132.
Grosjean, D., Miguel, AH, Tavares, TM, 1990. Urban air pollution in Brazil: acetaldehyde and other carbonyls. Atmos. Environ. B Urban Atmos. 24, 101–106. https://doi.org/
10.1016/0957-1272(90)90015-M.
Grosjean, D., Grosjean, E., Moreira, LFR, 2002. Speciated ambient carbonyls in Rio de Janeiro, Brazil. Environ. Sci. Technol. 36, 1389–1395. https://doi.org/
10.1021/es0111232.
Hajizadeh, Y., Mokhtari, M., Faraji, M., Mohammadi, A., Nemati, S., Ghanbari, R., Abdolahnejad, A., Fard, R.F., Nikoonahad, A., Jafari, N., Miri , M., 2018. Trends of BTEX in the
central urban area of Iran: a preliminary study of photochemical ozone pollution and health risk assessment. Atmos. Pollut. Res. 9, 220–229. https:// doi.org/10.1016/
j.apr.2017.09.005.
Hall, D., Wu, C.-Y., Hsu, Y.-M., Stormer, J., Engling, G., Capeto, K., Wang, J., Brown, S., Li, H. -W., Yu, K.-M., 2012. PAHs, carbonyls, VOCs and PM2.5 emission factors
for pre-harvest burning of Florida sugarcane. Atmos. Environ. 55, 164–172. https://doi.org/10.1016/j.atmosenv.2012.03.034.
Hassan, SK, El-Abssawy, AA, Khoder, MI, 2012. Characteristics of gas–phase nitric acid and ammonium–nitrate– sulfate aerosol, and their gas–phase precursors in a suburban
area in Cairo, Egypt. Atmos. Pollut. Res. 4, 117–129. https://doi.org/10.5094/apr.2013.012.
IBGE - Brazilian Institute of Geography and Statistics, 2021. Estimates of the Population of Municipalities. https://www.ibge.gov.br/estatisticas/sociais/populacao/
9103-estimativas-de-populacao.html?=&t=results
´ (accessed August 25, 2023).
IMGA - Instituto Mineiro de Gest˜ ao das Aguas, 2018. Water Quality Monitoring. http://portalinfohidro.igam.mg.gov.br/monitoramento-de-qualidade-das-aguas (accessed August
25, 2023).
INMET - Brazilian National Institute of Meteorology, 2018. Station and Data. https://portal.inmet.gov.br/ (accessed August 25, 2023).
International Agency for Research on Cancer (IARC), 1987. Overall Evaluations of Carcinogenicity: An Update of IARC Monographs. WorldHeal. Organ. https://
monographs.iarc.fr/iarc-monographs-on-the-evaluation-of-carcinogenic-risks-to-humans-80/ (accessed August 25, 2023).
International Agency for Research on Cancer (IARC), 2004. Monographs on the Evaluation of Carcinogenic Risks to Humans Formaldehyde, 2-Butoxyethanol. IARC
Book http://monographs.iarc.fr/ENG/Monographs/vol62/ (accessed August 25, 2023).
Jion, MMMF, Jannat, JN, Mia, MY, Ali, MA, Islam, MS, Ibrahim, SM, Islam, ARMT, 2023. A critical review and perspective of NO2 and SO2 pollution over
Asia: hotspots, trends, and sources. Sci. Total Environ. 1, 162851 https://doi.org/10.1016/j.scitotenv.2023.162851.
Kim, K.-H., Jung, K., Jeon, E.-C., Shon, Z.-H., Kim, N.-J., Phan, N.-T., 2012. Analysis of ammonia variation in the urban atmosphere. Atmos. Environ. 65, 177–185. https://doi.org/
10.1016/j.atmosenv.2012.10.049.
Kro, S., Namies, J., 2010. Monitoring VOCs in atmospheric air I. On-line gas analyzers. Trends Anal. Chem. 29 (9), 1092–1100. https://doi.org/10.1016/j.
trac.2010.05.007.
Lan, TTN, Nishimura, R., Tsujino, Y., Imamura, K., Warashina, M., Hoang, NT, Maed, Y., 2004. Atmospheric concentrations of sulfur dioxide, nitrogen oxides, ammonia, hydrogen
chloride, nitric acid, formic and acetic cids in the south of Vietnam measured by the passive sampling method. Anal. Sci. 20, 213–217. https://doi.org/10.2116/analsci.20.213.

Liu, Yanjun, Lu, W., Guo, H., Ming, Z., Wang, C., Xu, S., Liu, Yanting, Wang, H., 2016. Aromatic compound emissions from municipal solid waste landfill: emission
factors and their impact on air pollution. Atmos. Environ. 139, 205–213. https://doi.org/10.1016/j.atmosenv.2016.05.043.
Liu, H., Alharthi, M., Atil, A., Zafar, MW, Khan, I., 2022. A non-linear analysis of the impacts of natural resources and education on environmental quality: green
energy and its role in the future. Policy 79, 102940. https://doi.org/10.1016/j.resourpol.2022.102940.
Manco, A., Ciccioli, P., Famulari, D., Brilli, F., Ciccioli, P., Di Tommasi, P., Magliulo, V., 2022. Real-time air concentrations and turbulent fluxes of volatile organic compounds
(VOCs) over historic closed landfills to assess their potential environmental impact. Environ. Pollut. 309, 119748 https://doi.org/10.1016/j. envpol.2022.119748.

Mattila, J.M., Brophy, P., Kirkland, J., Hall, S., Ullmann, K., Fischer, EV, Brown, S., McDuffie, E., Tevlin, A., Farmer, D.K., 2018. Tropospheric sources and sinks of
gas-phase acids in the Colorado front range. Atmos. Chem Phys ´ 18, 12315–12327. https://doi.org/10.5194/acp-18-12315-2018.
Montero, L., Vasconcellos, PC, Souza, SR, Pires, MAF, Sanchez-Ccoyllo, OR, Andrade, MF, Carvalho, LRF, 2001. Measurements of atmospheric carboxylic acids and carbonyl
compounds in S˜ ao Paulo City, Brazil. Environ. Sci. Technol. 35, 3071–3081. https://doi.org/10.1021/es001875g.
Mraihi, R., Harizi, R., Mraihi, T., Taoufik Bouzidi, M., 2015. Urban air pollution and urban daily mobility in large Tunisia's cities. Renew. Subst. Energy Rev. 43,
315–320. https://doi.org/10.1016/j.rser.2014.11.022.
Mrowiec, B., 2014. Toluene in sewage and sludge in wastewater treatment plants. Water Sci. Technol. 69 (1), 128–134. https://doi.org/10.2166/wst.2013.563.
Muttray, A., Gosepath, J., Brieger, J., Faldum, A., Pribisz, A., Mayer-Popken, O., Jung, D., Roßbach, B., Mann, W., Letzel, S. ., 2009. No acute effects of an exposure to
50 ppm acetaldehyde on the upper airways. Int. Arch. Occup. Environ. Health 82, 481–488. https://doi.org/10.1007/s00420-008-0354-9.
National Institute for Occupational Safety and Health (NIOSH), 2019. Workplace Safety and Health Topics: Ammonia. https://www.cdc.gov/niosh/topics/ammonia/
default.html#:~:text=High%20levels%20of%20ammonia%20canduration%2C%20and%20work%20being%20done (accessed August 25, 2023).
National Pollutant Inventory (NPI), 2014. Ammonia (Total). http://www.npi.gov.au/resource/ammonia-total (accessed August 25, 2023).
National Toxicology Program (NPT), 2011. Report on Carcinogens, Twelfth ed.. In: http://ntp.niehs.nih.gov/ntp/roc/twelfth/reviewprocess.pdf (accessed August 25,
2023).
Neckel, A., Oliveira, MLS, Maculan, LS, Bodah, BW, Gonçalves Jr., AC, Silva, LFO, 2023. Air pollution in central European capital (Budapest) via self-made
passive samplers and sentinel-3B SYN satellite images. Urban Climate. 47, 101384 https://doi.org/10.1016/j.uclim.2022.101384.
NOAA - National Oceanic and Atmospheric Administration (NOAA), 2023. HYSPLIT Trajectory Model. https://www.ready.noaa.gov/hypub-bin/trajtype.pl (accessed
August 25, 2023).
Nogueira, T., Dominutti, PA, de Carvalho, LRF, Fornaro, A., Andrade, MF, 2014. Formaldehyde and acetaldehyde measurements in urban atmosphere impacted by the use of
ethanol biofuel: metropolitan area of Sao Paulo (MASP) 2012– 2013. Fuel 134, 505–513. https://doi.org/10.1016/j.fuel.2014.05.091.

16
Machine Translated by Google

FO Campos da Rocha et al. Urban Climate 52 (2023) 101739

Nojiri, R., Osada, K., Kurosaki, Y., Matsuoka, M., Sadanaga, Y., 2022. Variations in gaseous nitric acid concentrations at Tottori, Japan: long-range transport from the Asian
continent and local production. Atmos. Environ. 274, 118988 https://doi.org/10.1016/j.atmosenv.2022.118988.
Nolte, C.G., Solomon, P.A., Fall, T., Salmon, L.G., Cass, G.R., 1997. Seasonal and spatial characteristics of formic and acetic acids concentrations in the Southern
California atmosphere. Environ. Sci. Technol. 31, 2547–2553. https://doi.org/10.1021/es960954i.
OME - Ontario Ministry of the Environment, 2020. Ambient Air Quality Criteria. https://files.ontario.ca/mecp-ambient-air-quality-criteria-list-en-2020-05-01.pdf
(accessed August 25, 2023).
Omidvarborna, H., Baawain, M., Al-Mamun, A., 2018. Ambient air quality and exposure assessment study of the Gulf cooperation council countries: a critical review.
Sci. Total Environ. 636, 437–448. https://doi.org/10.1016/j.scitotenv.2018.04.296.
Padappayil, R.P., Borger, J., 2023. Ammonia Toxicity. National Library of Medicine. National Center for Biotechnology Information. https://www.ncbi.nlm.nih.gov/books/NBK546677/
(accessed August 25, 2023).
Portuguese Environmental Agency Air Quality Information (QualAR), 2021. National Database on Air Quality in Portugal. https://qualar.apambiente.pt/en/
downloads (accessed August 25, 2023). ˆ
Rocha, FOC, Campos, VP, 2023. AnaliseAr: passive demonstration device for simultaneous low-cost medications of 16 gaseous atmospheric pollutants/pollutants. Eng.
Sanit. Ambient 28, e20220138 https://doi.org/10.1590/S1413-415220220138.
Rodrigues, MC, Guarieiro, LL, Cardoso, MP, Carvalho, LS, da Rocha, GO, De Andrade, JB, 2012. Acetaldehyde and formaldehyde concentrations from sites impacted by heavy-
duty diesel vehicles and their correlation with the fuel composition: diesel and diesel/biodiesel blends. Fuel 92, 258–263. https://doi.org/ 10.1016/j.fuel.2011.07.023.

Salameh, T., Sauvage, S., Locoge, N., Gauduin, J., Perrussel, O., Borbon, A., 2019. Spatial and temporal variability of BTEX in Paris megacity: two-wheelers as a major
driver. Atmos. Environ. X 1, 100003. https://doi.org/10.1016/j.aeaoa.2018.100003.
Santana, FO, Campos, VP, Cruz, LPS, Luz, SR, 2017. Formaldehyde and acetaldehyde in the atmosphere of Salvador-Ba, Brazil, using passive sampling.
Microchem. J. 134, 78–86. https://doi.org/10.1016/j.microc.2017.04.032.
Santana, FO, Campos, VP, Santos, IF, Cruz, LP, Brito, AVS, 2019. Seasonal chemometric study of formaldehyde and acetaldehyde atmospheric levels and health
risk assessment, in urban areas of Salvador-Bahia, Brazil. Microchem. J. 147, 524–531. https://doi.org/10.1016/j.microc.2019.03.069.
Schuch, D., de Freitas, ED, Espinosa, SI, et al., 2019. A two decades study on ozone variability and trend over the main urban areas of the S˜ ao Paulo state, Brazil.
Environ. Sci. Pollut. Res. 26, 31699–31716. https://doi.org/10.1007/s11356-019-06200-z.
Seinfeld, JH, Pandis, SN, 2016. Atmospheric Chemistry and Physics: From Air Pollution to Climate Change, third ed. Wiley, New Jersey.
SENATRAN - National Traffic Deparment Brazil, 2020. National Fleet (2018). https://www.gov.br/infraestrutura/pt-br/assuntos/transito/conteudo-Senatran/frota-
of-vehicles-2018 (accessed August 25, 2023).
Shi, C., Saiz-Lopez, A., Gao, S., Fu, Q., Cui, H., Nan, J., Wang, S., Zhou, B., Wang, D., 2015. Atmospheric ammonia and its impacts on regional air quality over the
megacity of Shanghai, China. Sci Rep 5, 1–13. https://doi.org/10.1038/srep15842.
SMVKU- Senatsverwaltung für Mobilit¨ at, Verkehr, Klimaschutz und Umwelt, 2022. https://www.berlin.de/sen/uvk/umwelt/luft/luftqualitaet/luftdaten-archiv/
#jahr2022 (accessed August 25, 2023).
Souza, SR, Carvalho, LRF, 2001. Seasonality influence in the distribution of formic and acetic acids in the urban atmosphere of Sao ˜ Paulo City, Brazil. J. Braz.
Chem Soc 12, 755–762. https://doi.org/10.1590/S0103-50532001000600011.
Squizzato, R., Nogueira, T., Martins, L.D., et al., 2021. Beyond megacities: tracking air pollution from urban areas and biomass burning in Brazil. Climate Atmos. Sci. 4, 17. https://
doi.org/10.1038/s41612-021-00173-y.
Stern, A.C., 2015. Sources of Air Pollution and their Control: Air Pollution. Elsevier.
Sun, K., Tao, L., Miller, DJ, Pan, D., Golston, LM, Zondlo, MA, Griffin, RJ, Wallace, HW, Leong, YJ, Yang, MM, Zhang, Y., Mauzerall, DL, Zhu, T., 2017.
Vehicle emissions as an important urban ammonia source in the United States and China. Environ. Sci. Technol. 51, 2472–2481. https://doi.org/10.1021/acs. est.6b02805.

TCEQ - Texas Commission on Environmental Quality, 2016. Effects Screening Levels (ESLs). https://www.tceq.texas.gov/toxicology/esl/list_main.html (accessed
August 25, 2023).
Teng, X., Hu, Q., Zhang, L., Qi, J., Shi, J. , Xie, H., Gao, H., Yao, an urban environment in northern
China during wintertime. Environ. Sci. Technol. 51, 6839–6848. https://doi.org/10.1021/acs.est.7b00328.
UN-Habitat - United Nations Human Settlements Programme, 2022. World Cities Report 2022. https://unhabitat.org/wcr/#downloads (accessed August 25, 2023).
United States Environmental Protection Agency (USEPA), 2017a. Air Topics (Accessed August 25, 2023). https://www.epa.gov/environmental-topics/air-topics
(accessed August 25, 2023).
United States Environmental Protection Agency (USEPA), 2017b. Nitrogen Dioxide (NO2) Pollution, Nitrogen Dioxide Pollut. https://www.epa.gov/no2-pollution/
basic-information-about-no2#Effects (accessed August 25, 2023).
United States Environmental Protection Agency (USEPA), 2022a. Health Effects Notebook for Hazardous Air Pollutants: Formaldehyde. https://www.epa.gov/sites/
default/files/2016-09/documents/formaldehyde.pdf (accessed August 25, 2023).
United States Environmental Protection Agency (USEPA), 2022b. Health Effects Notebook for Hazardous Air Pollutants: Acetaldehyde. https://www.epa.gov/sites/
default/files/2016-09/documents/acetaldehyde.pdf (accessed August 25, 2023).
Vieira-Filho, MS, Ito, DT, Pedrotti, JJ, Coelho, LHG, Fornaro, A., 2016. Gas-phase ammonia and water-soluble ions in particulate matter analysis in an urban
vehicular tunnel. Environ. Sci. Pollut. Res. 23, 19876–19886. https://doi.org/10.1007/s11356-016-7177-0.
Wang, YT, Thomas, DD, 2017. Nitrogen oxides and their roles in cancer etiology. Curr. Pharmacol. Rep. 3, 151–161. https://doi.org/10.1007/s40495-017-0092-3.
World Health Organization - WHO, 2000. Air Quality Guidelines for Europe, 2nd ed. WHO Regional Publications, Copenhagen. European Series, no. 91. https://www.
euro.who.int/__data/assets/pdf_file/0017/123056/AQG2ndEd_5_2benzene.pdf. (accessed August 25, 2023).
World Health Organization - WHO, 2021. WHO Global Air Quality Guidelines: Particulate Matter (PM2.5 and PM10), Ozone, Nitrogen Dioxide, Sulfur Dioxide and
Carbon Monoxide. https://www.who.int/publications/i/item/9789240034228?ua=1 (accessed August 25, 2023).
World Health Organization - WHO, 2022. Ambient (Outdoor) Air Pollution. https://www.who.int/news-room/fact-sheets/detail/ambient-(outdoor)-air-quality-and-
health#:~:text=Ambient%20(outdoor)%20air%20pollution% 20in,and%20respiratory%20disease%2C%20and%20cancers (accessed August 25, 2023).
World Health Organization (WHO), 2003. Hydrogen Sulfide: Human Health Aspects. https://apps.who.int/iris/bitstream/handle/10665/42638/9241530537.pdf
(accessed August 25, 2023).
Yang, Z., Cheng, HR, Wang, ZW, Peng, J., Zhu, JX, Lyu, XP, Guo, H., 2019. Chemical characteristics of atmospheric carbonyl compounds and source identification of
formaldehyde in Wuhan, Central China. Atmos. Res. 228, 95–106. https://doi.org/10.1016/j.atmosres.2019.05.020.
Yli-Pelkonena, V., Viippola, V., Kotze, DJ, Set¨ ala, ¨ H., 2017. Greenbelts do not reduce NO2 concentrations in near-road environments. Urban Climate. 21, 306–317,
2017. https://doi.org/10.1016/j.uclim.2017.08.005.
Zscheischler, J., Martius, O., Westra, S., Bevacqua, E., Raymond, C., Horton, RM, Hurk, B., AghaKouchak, A., J´ez´equel, A., Mahecha, MD, Maraun, D., Ramos, AM, Ridder, NN, Thiery, W., Vignotto, E., 2020. A typology
of compound weather and climate events. Nat. Rev. Earth Environ. 1, 333–347. https://doi.org/10.1038/ s43017-020-0060-z.

17