art salamanca fuel 90 2011

8/3/2019 Art Salamanca Fuel 90 2011

1/12

Air quality assessment in a highly industrialized area of Mexico: Concentrationsand sources of volatile organic compounds

Elizabeth Vega, Gabriela Snchez-Reyna , Virginia Mora-Perdomo, Gustavo Sosa Iglesias, Jos Luis Arriaga,Teresa Limn-Snchez, Sergio Escalona-Segura, Eugenio Gonzalez-Avalos

Instituto Mexicano del Petrleo, Eje Central Lzaro Cardenas 152, Col. San Bartolo Atepehuacan, Distrito Federal C.P. 07730, Mexico

a r t i c l e i n f o

Article history:

Received 24 August 2010

Received in revised form 29 March 2011

Accepted 31 March 2011

Available online 24 April 2011

Keywords:

VOCs

Air quality

Industrial areas

Halocarbons

Mexico

a b s t r a c t

Parallel to the economical benefits brought by the oil industry in Mexico, there have been some negative

environmental effects due to emission of pollutants to the atmosphere. Salamanca, a city located inside

one of the most important industrial corridors of the country, has been frequently affected by elevated

concentrations of sulfur dioxide and particle matter. However, little is known about volatile organic com-

pounds (VOCs), which in this study are analyzed along with criteria pollutants and meteorological

parameters during FebruaryMarch 2003 at urban, suburban and rural sites. Although sulfur dioxide

average levels were $0.017 ppm, a high concentration event ($0.600 ppm), attributable to emissions

from the oil refinery and the thermoelectric power plant, was observed at the urban site at night time.

The VOCs concentration varied from 170 50 ppbC (rural) to 699 212 (urban) and were constituted

by 40% alkanes, 13% aromatics, 11% olefins and 11% of halogenated. The most abundant species were pro-

pane (167 40 ppbC), n-butane (91 23 ppbC), toluene (51 10 ppbC) and i-pentane (44 7 ppbC), that

are related to combustion processes. Freon-114, methyl bromide and 1,2-dichloroethane which are likely

emitted by application of pesticides, soil fumigation and fabrication of chemicals, showed high concen-

trations (48 10, 50 10 and 32 6 ppbC respectively) in the rural sites, highlighting the importance

of control measurements implementation for these species, as they represent a potential hazard for pub-

lic health. Moreover, these halocarbons showed similar ratios regardless the monitoring site, suggesting

same source. Modeling results indicated that meteorological conditions generally transport air masses to

the northeast rural areas where the highest concentrations of ozone were calculated.

2011 Elsevier Ltd. All rights reserved.

1. Introduction

Over the last century, the oil industry has emerged as the pri-

mary energy source [1]. Currently, the life style of human societies

depends on energy (electricity generation, natural gas, crude oil

and its refined products, coal, etc.); without it, societies as we

know them would collapse. Even though the oil industry has made

important contributions to the global economy, usually this has

been accompanied with negative environmental impacts from avariety of activities such as oil drilling, refinery, oil spillage, gas

and flaring. Moreover, deterioration of the environment may not

be circumscribed to the local scale, it can reach regional and global

extent due to the emission of precursors of secondary pollutants

and chemical species that contribute enhancing global warming

and stratospheric ozone depletion [24]. Public health may also

be affected if emissions contain toxic or carcinogenic species [5].

In recent years concern over public health and environmental pro-

tection has become a critical issue, this means that a growing

amount of investment and effort is dedicated to reconcile the envi-

ronment and development of countries.

The economy of Mexico strongly depends on oil industry; in

2005 the primary distillation capacity (1540 MBD) ranked the

country on the top 15 worldwide and 4 in Latin America. The crude

is processed in 6 oil refineries which mainly produce gasoline, die-

sel, jet fuel, coal, asphalt, and lubricants. The third most important

refinery is the Ing. Antonio M. Amor, which processes 197 MBD [6].

The refinery and a variety of industries constitute one of the mostimportant industrial corridors of Mexico, known as the Bajo Indus-

trial Corridor (BIC) located in the State of Guanajuato, in the central

area of the country (Fig. 1). The BIC has nearly 465 industries, from

medium to large size, including Chemical, Power Generation, Food

Processing, Textile and Metal-mechanic [7].

The Salamanca city, with 250,000 inhabitants, is located at

203400900N latitude and 1011105100W longitude, at 1720 m above

mean sea level [8]. The Salamanca municipality encompasses a

total area of 774 km2. The agriculture is now the second most

important economical activity, with a designated area of about

80% of the municipality. The impact of agriculture on the environ-

ment is important, especially for the use of fertilizers, pesticides,

0016-2361/$ - see front matter 2011 Elsevier Ltd. All rights reserved.doi:10.1016/j.fuel.2011.03.050

Corresponding author. Tel.: +52 55 91757558.

E-mail address: [email protected] (G. Snchez-Reyna).

Fuel 90 (2011) 35093520

Contents lists available at ScienceDirect

Fuel

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / f u e l
http://dx.doi.org/10.1016/j.fuel.2011.03.050mailto:[email protected]://dx.doi.org/10.1016/j.fuel.2011.03.050http://www.sciencedirect.com/science/journal/00162361http://www.elsevier.com/locate/fuelhttp://www.elsevier.com/locate/fuelhttp://www.sciencedirect.com/science/journal/00162361http://dx.doi.org/10.1016/j.fuel.2011.03.050mailto:[email protected]://dx.doi.org/10.1016/j.fuel.2011.03.050


2/12

deforestation in the uplands and post-harvesting burning. The big-

gest impact to the atmosphere is represented by the post-harvest-ing burning, due to the emission of large amounts of ozone

precursors and particulate matter [8].

According to the 2000 BIC Emissions Inventory, the emissions of

particles (PM10), sulfur dioxide (SO2), carbon monoxide (CO), nitro-

gen oxides (NOX) and hydrocarbons (HC) were 71,443, 112,480,

1,650,772, 142,183 and 260,296 tons per year, respectively. It is

estimated that PM10 are released to the atmosphere mainly by

commercial and service activities, SO2 and NOX by electricity gen-

eration, and CO and hydrocarbons by vehicle exhaust. Salamanca

contributes with 18% of PM10, 92% of SO2, 8% of CO and HC and

14% of NOX of the total BIC emissions. Thus, Salamanca is by far

the major generator of SO2 emissions in the region; while other cit-

ies release ozone precursors [8].

Parallel to the economical development of the BIC, there havebeen some adverse environmental impacts which have brought

the attention of government agencies, civil and private associa-

tions. As a result, since 2000 a Monitoring Network in Salamanca

routinely measures CO, SO2, NO2, O3, and PM10. According to local

environmental authorities, the SO2 air quality standard, AQS,

(0.13 ppm in a 24 h average, no more than once per year) was ex-

ceeded 13%, 24% and 22% of days in 2000, 2001 and 2002 respec-

tively in downtown Salamanca, mainly during winter. The

exceedences of other gaseous pollutants is less frequent, for in-

stance, NO2 and CO have been practically below their AQS

(0.21 ppm, 1 h average, and 11 ppm in 8 h average respectively)

[8].

Although total mass of criteria pollutants is routinely measured,

little is known in this highly industrialized region about the gas-eous and particle contaminants that are not included in the local

monitoring network, such as the volatile organic compounds

(VOCs). The negative effects of VOCs on the environment and pub-

lic health are well documented. From the environmental point of

view, some VOCs (i.e. olefins and aromatics which are mainly

anthropogenic) are reactive species that break out the natural

equilibrium of generationdestruction of tropospheric ozone, thus

the concentrations of this compound and other photochemically-

produced pollutants are frequently high in the urban environment.Besides, the reactive organic gases can partition into the aerosol

phase generating secondary organic aerosols. Other important

group of VOCs is constituted by the halogenated species, which

are originated almost exclusively from anthropogenic emissions

due to its usage as an industrial solvent and degreaser. Some of

these compounds have been the focus of intensive research, such

as the chlorofluorocarbons due to their participation in the strato-

spheric ozone depletion. In addition, many of the halogenated spe-

cies represent a potential hazard to human health due to the toxic

and/or carcinogenic effect [915].

The public opinion on air quality deterioration in Salamanca,

encouraged PEMEX (National Oil Company) to support an exten-

sive 2-week monitoring field study with the aim of augmenting

the knowledge of sources, transport and fate of air pollutants in

the region, therefore effective control measurements of atmo-

spheric pollution can be designed. The main findings of such cam-

paign are presented in this work, particularly the chemical

characterization, distribution and origin of VOCs, as well as the

meteorological parameters that influence the dilution and trans-

port of pollutants. The later was also estimated by applying a 3 D

air quality model.

2. Field campaign

As an outcome of a collaborative effort, the Instituto Mexicano

del Petroleo (IMP), the Instituto de Ecologa de Guanajuato, the

Instituto de Investigaciones Cientficas at the Universidad de Guan-

ajuato, the Centro de Ciencias de la Atmsfera at the UniversidadNacional Autnoma de Mxico, the Patronato de Salamanca, and

the Ing. Antonio M. Amor Oil Refinery, with the PEMEX sponsor-

ship, carried out a field monitoring campaign, from February 21

to March 9 2003. The main objectives were to chemically charac-

terize the air pollution in the urban area of Salamanca in both

the particle and gas phases, and to assess the potential impact of

pollutants in the regional scale. The interested reader can consult

Vega et al. [16] for the particulate matter results found in this

region.

The first week of measurements was focused on the character-

ization of urban air quality; while the second week was designed to

evaluate the regional impact of urban emissions. The monitoring

sites of the urban domain (10 10 km) and of the regional domain

(80 80 km) are shown in Fig. 2A and B. Table 1 shows site loca-tion, description, sampling period and measurements performed.

Three automated samplers (VOCCS-ANDERSEN and AVOCS-

ANDERSEN models) with a Viton diaphragm pump were used to

collect VOCs (defined in this work as hydrocarbons from C2 to

C12) in canisters over 12 h period (06001800 and 18000600 local

time) in the urban sites and 24 h period in the rural/boundary sites.

A total of 80 canisters were analyzed in the Laboratory by cryo-

genic pre-concentration/high-resolution GC technique, similar to

the TO-14A protocol [17].

Water Sep-Pak DNPH-Silica cartridges were used to trap car-

bonyl species. Twelve samples were taken during the first week

of the campaign at Cruz Roja (CR) urban site from 0600 to 0900

and from 1200 to 1500. The derivatives were eluted and analyzed

by HPLC with UV photodiode array detector according to the TO-11A protocol [18]. Criteria pollutants were measured using a

GuanajuatoState

Salamanca

County

(B)

Zacatecas

Sn. Luis Potos

Jalisco

Michoacn

Quertaro

Len

Irapuato

Salamanca

Celaya

(A)

Fig. 1. Geographical location of: (A) Salamanca County and (B) The Bajo Industrial

Corridor.

3510 E. Vega et al./ Fuel 90 (2011) 35093520


3/12

mobile Lab equipped with conventional analyzers (Monitor Labs).

Methods used to determine these pollutants were NOM-034

SEMARNAT 1993 using dispersive spectroscopy for CO (detectionlimits from 0 to 50 ppm); NOM-037 SEMARNAT 1993 using

quimioluminescence for NOx (detection limits from 0 to

0.50 ppm); USEPA EQOA 0193 091 using UV photometry

for O3 (detection limits from 0 to 1.0 ppm) and USEPA EQSA

0193 092 using pulse fluorescence for SO2 with detection limits

from 0 to 1.0 ppm.

Along the field campaign, the surface meteorological parame-

ters temperature (T), relative humidity (RH), wind speed (WS),

wind direction (WD), atmospheric pressure (P) and solar radiation

(SR) were also measured at three sites in the urban area and at four

in the boundary sites. The vertical thermodynamic profile variables

(P, T, RH and horizontal wind vector) were measured using a Dig-

icora II radiosonde system from Vaisala (Model SPS-220). Three

radiosondes were launched every day at 0800, 1200 and 1800 atSI, JR and VS sites. The information was used to determine the mix-

ing height based on potential temperature and specific humidity

profiles, and also as input for the mesoscale meteorological model

RAMS [19] and the 3D air quality model [20].

3. Results

3.1. Criteria pollutants

The average and standard deviation for 1 h average concentra-

tion of criteria pollutants (CO, NO2, SO2 and O3) and nitrogen oxi-

des (NOX) at urban (CR and DIF), suburban (CG and US) and rural

(MI, JR and VS) sites are shown in Table 2. As expected, the atmo-

spheric concentrations of CO, NO2 and NOX were considerably

higher at the urban sites compared with the suburban and rural

ones. The suburban sites are surrounded by crop fields and un-

paved small roads with little traffic of heavy-duty vehicles which

may influence the measurements, however it is expected to beminimal.

(A)

(B)

-101.30 -101.25 -101.20 -101.15 -101.10 -101.05 -101.00

20.50

20.55

20.60

20.65

20.70

CG

VAUS

CA

CR

NADIF

RT

-101.6 -101.5 -101.4 -101.3 -101.2 -101.1 -101.0 -100.9 -100.8 -100.7

20.1

20.2

20.3

20.4

20.5

20.6

20.7

20.8

20.9

21.0

VS

PN

JR

SI

MI

SALSAL

Fig. 2. Monitoring sites location: (A) Urban Area: Cruz Roja (CR) and DIF; suburban: Crdenas (CA), Cerro Gordo (CG), Universidad la Salle (US), and Valtierrilla (VA); the

Refinery (R) and the Power Plant (T). (B) Regional Area: Silao (SI), Mirandas (MI), Juventino Rosas (JR), Pueblo Nuevo (PN) and Valle de Santiago (VS), SAL is the Salamanca City.

E. Vega et al. / Fuel 90 (2011) 35093520 3511


4/12

Regarding ozone, it is noticeable that average levels in the urban

sites were 0.022 ppm with maximum of 0.094 ppm, while the rural

sites reached 0.031 ppm in average and 0.114 ppm maximum. The

diurnal variations of this oxidant at CR (urban), suburban (CG) and

rural (MI and VS) have been plotted in Fig. 3. For all sites, maxi-

mum levels took place from 1200 to 1500, being about 40% higher

in the rural site (MI) than the urban and suburban ones. The mon-

itoring site of VS located within a natural protected area, 25 km

south of the city at an altitude 300 m above the level of Salamanca,

registered ozone average concentration of 0.038 ppm and maxi-

mum of 0.052 which could be considered as background concen-

trations. These results are within the range of ozone levels

reported at rural areas of Canada [21]. In addition, it is known that

areas of high altitude (the study region is 1720 m above sea level)

could be influenced by inputs from free troposphere [2123]. As it

is discussed later, the peak of mixing height was observed every-

day at 1800 reaching 3500 m above ground level, making possible

this phenomenon, although more measurements would be neces-

sary to reach a conclusion.

The highest ozone concentrations were measured at rural MI

which is located 6.5 km north of the city. To investigate the influ-

ence of surface wind direction and speed on the spatial distributionof ozone, an analysis was performed dividing the data-set into four

subsets according to surface wind direction: north (31545), east

(45135), south (135225) and west (225315). As expected,

the results reflected the influence of site location regarding main

pollution sources. The concentrations of ozone at MI were higher

when the wind blew from the west and south, i.e., the site was

downwind major pollution sources. These wind directions were

observed $64% of the time, mostly from 1200 to 1900, which sug-

gest that ozone levels at MI were strongly influenced by trans-

ported ozone. Moreover, the correlation analysis indicated that

levels observed at MI and CR are associated during this meteoro-

logical condition (R2 = 0.72). The same analysis was performed

using data of the opposite wind direction (north), showing an even

stronger relationship (R2

= 0.91) between the urban and ruralozone levels. It has to be added that north winds generally blew

Table 1

Location and type of monitoring sites during the monitoring field Campaign in Salamanca, Mexico. FebruaryMarch 2003.

Site name, code and

coordinates

Site type Site description Measurement

period

Measurements

Cruz Roja (CR) 20.58N,

101.20W

Urban Located in the main street of Salamanca with high vehicular (light and duty) traffic all

day; 2 km west of the Refinery and the Power Plant

February 21

March 9 2003

Criteria pollutantsa,

VOCs, carbonyl

species

DIF 20.56N, 101.20W Urban Located in a residential area with high vehicular traffic, mainly of gasoline-powered

vehicles

February 21

March 9 2003

Criteria pollutantsa

Crdenas (CA) 20.63N,

101.22W

Suburban The site is in a crop field near an unpaved road with little vehicular traffic February 22

28 2003


Cerro Gordo (CG)

20.59N, 101.13W

Suburban 7 km southwest of the Refinery and Power Plant, 200 m of a highway used mainly by

heavy-duty vehicles. The site is surrounded by crop fields

February 21

28 2003


VOCs

Valtierrilla (VA)

20.56N, 101.13W

Suburban Located in a populated community with moderate transit of gasoline and diesel

vehicles

February 22

28 2003


Universidad La Salle

(US) 20.55N,

101.23W

Suburban Located inside the University, southwest Salamanca; site is surrounded by vegetation;

400 m away from a four-lane road with light and duty-vehicles traffic

February 21

28 2003


VOCs

Silao (SI) 20.59N,

101.42W

Rural/

boundary

The site was used for meteorological measurements, located northwest Salamanca March 29

2003

Radiosondes

Valle de Santiago (VS)

20.35N, 101.20W

Rural/

boundary

Located inside a natural protected area known as Siete Luminarias. The site is 300 m

above the Salamanca level, 1.5 km away from a road with little vehicular traffic

March 29

2003


radiosondes

Pueblo Nuevo (PN)

20.55N, 101.35W

Rural/

boundary

The site is surrounded by crop fields near one unpaved road with little light-duty

vehicular traffic

March 29

2003


radiosondes

Juventino Rosas (JR)

20.64

N, 101.00

W

Rural/

boundary

22 km northeast Salamanca, in a commercial area with buildings 46 m height with

moderated vehicular traffic

March 19

2003


VOCs, radiosondesMirandas (MI)

20.56N, 101.14W

Rural/

boundary

This monitoring site is located 6.5 km north Salamanca, in an wide-open area

surrounded by crop fields

March 19

2003


VOCs

a Criteria Pollutants were measured each minute and include O3, CO, NO2, SO2 and PM10.

Table 2

Basic statistics of concentrations of pollutants (ppm) in urban (CR and DIF), suburban (CG and US) and rural (JR, VS and MI) sites of Salamanca.

Species Urban Suburban Rural

Average Maximum Minimum SD n Average Maximum Minimum SD n Average Maximum Minimum SD n

O3 0.022 0.094 0.001 0.018 699 0.021 0.072 0.001 0.017 353 0.034 0.114 0.001 0.020 525

CO 2.200 19.500 0.960 1.150 699 0.158 1.890 0.001 0.240 347 0.038 0.570 0.001 0.067 525

SO2 0.017 0.310 0.001 0.043 699 0.016 0.269 0.001 0.034 294 0.011 0.318 0.001 0.026 525

NO2 0.022 0.064 0.002 0.013 699 0.014 0.047 0.002 0.008 353 0.010 0.046 0.001 0.008 525

NOX 0.055 0.425 0.002 0.045 699 0.029 0.127 0.001 0.028 344 0.014 0.092 0.001 0.013 525

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0 3 6 9 12 15 18 21

CR MI CG VS

O3(ppm)

Fig. 3. Hourly average ozone concentration (ppm) at urban (CR), suburban (CG) andrural (MI and VS) sites of Salamanca during February 21March 8, 2003.

3512 E. Vega et al./ Fuel 90 (2011) 35093520


5/12

from 0000 to 0700 and 2100 to 2300, thus measurements indicate

background concentrations. The same analysis was performed for

urban CR and rural VS; the data-set was filtered so only measure-

ments taken when the rural site was located downwind Salaman-

ca. A correlation coefficient of 0.55 was obtained between these

sites (the site is 25 km south), while similar coefficient was ob-

tained for the analysis using only winds from the south

(R2

= 0.57). In summary, the results suggest that transport of airmasses from downtown Salamanca and the elevation of the study

area play an important role in the high ozone concentrations ob-

served in the rural sites.

To evaluate the effect of wind direction on the concentration of

the other pollutants, the average and standard deviation of concen-

trations of criteria pollutants and nitrogen oxides were calculated

for each of the wind sectors above described. At CR, carbon monox-

ide and nitrogen dioxide registered higher levels when the wind

blew from the south or east, which is explained by the location

of this site in the northwest of Salamanca. On the other hand,

ozone levels at CR were statistically lower when the wind blew

from the north and higher when the wind blew from east. The con-

centrations of SO2 showed higher values associated with east-

winds (where the refinery is located) and lower for any other wind

direction. At suburban CG, carbon monoxide and nitrogen dioxide

were higher when the wind blew from the east or south, as it was

envisaged. The concentrations of pollutants were higher when the

wind came from the south at the rural site, which is explained by

the sites location (Table 3). Moreover, the observed results agree

with simulation results of transport of air masses, which are dis-

cussed in Section 3.4.

As mentioned before, the largest sources of SO2 in the study re-

gion are the Refinery and the Power Plant facilities, both located in-

side the Salamanca city. Even though emissions are released to the

atmosphere by elevated stacks (3060 m above the ground), the

natural atmospheric processes such as turbulence, thermal inver-

sions and high pressure systems, may increase the concentration

and residence time of pollutants. Fig. 4 exemplifies the typical daily

variation of SO2 concentration. From 0000 to 0600, due to stableatmospheric conditions, the highest concentrations were observed

at CR, located 2 km west of major emission sources (the other two

sites, MI, and JR, are 6.5 km north and 22 km northeast, respec-

tively). From 0600 to 0900 the wind blew from the east, therefore

CR site received directly the SO2 emissions. Around 0900, the sur-

face wind direction blew from SSW, remaining under these condi-

tions the rest of the day, as a result, concentrations at CR showed a

decrement while at MI an increment was registered. It was found

that short-term variability periods (e.g. at CR at 09000930 and

at MI at 12001220 in Fig. 4) were associated to wind speed lower

than 1.0 ms1. According to RAMS modeling results, surface winds

before 0900 were driven by cold air draining from the near moun-

tains towards lowest topographic levels in the basin, accumulating

air pollutants over the city. After 0900 and due to synoptic wind

forcing, the wind blew from SW in the whole area, sweeping pollu-

tants out of the city. The estimated results were validated againstmeteorological measurements finding good correspondence be-

tween observed and predicted values.

During the monitoring campaign, high SO2 concentrations

(>0.60 ppm, 1 min sampling-time) were registered once at CR at

0200; such high concentrations were associated with persistent

east wind direction (8295) and wind speed of 2.6 ms1, which

was higher than the average registered at this time of the night

(1.3 ms1). Moreover, concentrations above the average of toluene,

m,p-xylenes, benzene, Freon-114 and methyl chloroform were also

registered during this event, suggesting emissions from industrial

activities.

The night-time high SO2 concentration events were frequently

observed in 19992002 and were associated with venting activities

of the Refinery and/or Power Plant [8]. However since 2003 the airquality program began to operate in the region, tackling SO2 and

particle matter problems mostly [24]. The control measurements

include the usage of fuels with low sulfur content in the power

plant and an increment in the natural gas consumption thus is pos-

sible that the first positive results were observed during the field

campaign.

3.2. Volatile organic compounds

3.2.1. Concentrations

Approximately 200 chemical species of VOCs were identified

and quantified by GC analysis. A classification was made according

Table 3

Wind directional analysis for concentrations of criteria pollutants (ppm) during FebruaryMarch 2003 in Salamanca, Mexico.

Site Species Wind direction

North (31545) East (45135) South (135225) West (225315)

CR O3 0.016 0.021 0.022 0.021 0.018 0.017 0.021 0.019

CO 1.946 0.927 2.219 1.104 2.937 1.604 2.201 1.094

NO2 0.032 0.010 0.037 0.010 0.034 0.012 0.028 0.010

SO2 0.020 0.046 0.122 0.139 0.012 0.034 0.008 0.020

CG O3 0.024 0.021 0.012 0.012 0.021 0.018 0.020 0.019

CO 0.161 0.229 0.383 0.518 0.264 0.404 0.158 0.275

NO2 0.016 0.008 0.020 0.010 0.016 0.009 0.014 0.008

SO2 0.031 0.046 0.010 0.019 0.030 0.058 0.030 0.049

MI O3 0.016 0.009 0.015 0.010 0.029 0.025 0.044 0.021

CO 0.074 0.151 0.068 0.121 0.129 0.149 0.093 0.136

NO2 0.011 0.005 0.012 0.006 0.017 0.008 0.009 0.007

SO2 0.004 0.007 0.004 0.010 0.032 0.063 0.014 0.047

March 2, 2003

SO

2(ppm)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

00:00 03:00 06:00 09:00 12:00 15:00 18:00 21:00 24:00

E S-W

NNE S-SW

N SW N

erraticMI

JR

CR

Fig. 4. Time series for SO2 concentrations at CR, MI and JR during March 2, 2003.

Upper horizontal lines indicate the observed wind direction.

E. Vega et al. / Fuel 90 (2011) 35093520 3513


6/12

to the functional group, indicating that alkanes were the most

abundant group ($40% of the total mass), followed by aromatics

($13%), olefins ($11%) and halogenated compounds ($11%). The

total mass concentration for samples taken from 0600 to 1800

(diurnal samples) varied from 699 212 ppbC at the urban site

(CR) to 170 50 ppbC at the suburban site (US), which is located

southwest of the city. The concentrations of VOCs taken from

1800 to 0600 (nocturnal samples) were higher than those mea-sured in the morning, reaching 956 120 ppbC at CR and

554 95 ppbC at US. Moreover, diurnal samples showed different

composition compared to nocturnal samples: the proportion of al-

kanes was higher during the night at CR, while halogenated and

the unidentified groups showed higher percentages at night in

the suburban sites (Fig. 5A and B). The higher levels of VOCs during

the night period were mainly driven by the increase of emissions of

propane, n-butane, i-butane, Freon-114 and toluene. Velasco et al.

[25] reported that VOCs levels for the Mexico City Metropolitan

Area during 2003 fluctuated from 2473 in the industrial area to

83 ppbC in the rural site for morning samples and from 1467 to

81 ppbC for afternoon samples. A simple comparison between val-

ues of VOCs measured in Mexico City and Salamanca, even though

of the difference in sampling-time, indicated that concentrations

observed in Salamanca were similar with some sites of Mexico

City. Regarding the distribution of VOCs by type, it was observed

the same order of abundance (i.e. alkanes > aromatics > olefins),

although the proportion of olefins in Salamanca was twice the con-

centration reported for Mexico City.

The highest concentrations of individual VOCs measured in the

atmosphere of Salamanca during the sampling campaign are de-

scribed in Table 4. At the urban and suburban sites, propane was

the most abundant species, followed by n-butane, toluene and i-

pentane. The first two species are attributable to the wide usage

of LPG for cooking and heating; on the other hand, toluene and i-

pentane may be emitted by both mobile and industrial sources.

Propane and n-butane are reported as the most abundant species

in the atmosphere of Mexico City [25], showing concentrations

similar to those observed in this study.

The halogenated species such as Freon-114, methyl bromide,

1,2-dichloroethane, 1,2-dichloropropane and vinyl chloride

showed levels noticeably high at the suburban/rural sites, espe-

cially during the night time. The concentrations of these speciesare markedly higher than those observed in an industrial area of

China, where Freon-114 and methyl bromide are $16 and $18

pptv [26,27]. Therefore, the results found in this study highlight

the importance of carrying out continuous monitoring of VOCs,

so control measurements can be taken to reduce population

exposure.

Formaldehyde, acetaldehyde and acetone average concentra-

tions were 3.75 1.94, 2.45 1.45 and 7.71 6.38 ppb, respec-

tively. Formaldehyde may have a significant influence on the

local photochemistry, more than any other carbonyl species [28].

However, concentrations of formaldehyde were low in comparison

to those measured in the southwest and downtown Mexico City

(13.3 and 23.9 ppb respectively) [29,30].

3.2.2. Sources of VOCs

According to abundance and spatial distribution of VOCs species

in Salamanca, the following categories can be identified as the ma-

jor contributors:

(1) Vehicular Emissions (gasoline and diesel-powered vehicles).

The emissions of vehicles powered by gasoline and diesel are

highly loaded with ethene, acetylene, propene, MTBE, n/i-

pentane, 2,2-dimethylbutane, 2-methylpentane, benzene,

toluene, and xylenes [31]. These species were observed in

all monitoring sites in concentrations high enough to rank

them among the first 15 (see Table 4). MTBE average levels

were 4.79 3.34 ppbC at the urban site (not shown in Table

4), representing about 1.0% of the total VOCs. By comparison,a percentage lower than 2% is reported for samples collected

in Mexico City [25].

(2) LPG handling and leakage. This source is characterized by

emissions rich in propane, i-butane, and n-butane [32]. Pro-

pane was the major species at all sites (except rural MI).

(3) Refining processes. Emissions from refining processes are

composed by a large variety of species, which depend on

the process itself. It is reported in the literature that ethane,

propane, n/i-pentane, toluene and formaldehyde are emitted

by refining oil processes and by fugitive emissions [33,34].

(4) Other sources: The high levels of halogen species such as

methyl bromide, methyl chloride, 1,2-dichloropropane and

dichloroethane reveal the presence of sources related with

agriculture activities (e.g. application of pesticides and soilfumigants) and with industrial solvent and degrease pro-

cesses [35].

As mentioned previously, burning of agriculture debris has been

a frequent activity in the region that generates large amounts of

particles and gases, mainly CO and organics. The presence of aceto-

nitrile in ambient air indicates this source [36,37]. We have used

styrene as surrogate species for acetonitrile, since the former has

also been considered marker for biomass burning in the rural envi-

ronments [25]. Average concentrations of styrene at the urban site

were 2.0 0.88 ppbC, while concentrations of 2.55 2.27 and

1.10 0.80 ppbC were found at the rural sites MI and JR, respec-

tively. Together, concentrations of styrene and characteristics of

MI site (surrounded by crop fields), suggest that this site is influ-enced by biomass burning emissions. However, no clear relation-

39 38 3641

46

20

816

10

9

19

12

1614

13

4

9

7 1011

1

1

2 43

9

27

21 16 13

7 5 3 5 6

0

20

40

60

80

100

CR CG MI JR US

alkanes olefins aromatics halogenated

oxigenated unidentified HC2

%

(A)

5044

28

14

5

12

17

9

5

4

17

28

1

1 1

921 24

5 4 1

0

20

40

60

80

100

CR US

alkanes olefins aromatics halogenated

oxigenated unidentified HC2

%

(B)

Fig. 5. Distribution of abundance of VOCs in Salamanca. (A) Diurnal samples (06001800); (B) nocturnal samples (18000600).

3514 E. Vega et al./ Fuel 90 (2011) 35093520


7/12

ship was found between styrene and other species indicative of

biomass burning.

A series of scatter plots were constructed to examine the contri-

bution of urban and industrial emissions in the region of study.

Toluene and propene are emitted by both sources; while MTBE is

a distinctive vehicular emission tracer. Fig. 6A shows the scatter

plots for MTBE and toluene and Fig. 6B for MTBE and propene. Both

figures show a subset of data where the two species correlate lin-

early indicating the mobile source, while the second non-corre-

lated subset indicates an industrial origin. The sites mostly

influenced by mobile sources were US, MI, JR and CR, while CG

and in some days CR showed influence by industrial emissions.

This result agrees with the assumption that pollutants are mainly

transported from the city towards northeast, and that CR and CG

were the sites with the higher Refinery and Power Plant emissions

impact.

Propane and n-butane are known fingerprint species of leakages

and unburned LPG [32]. Fig. 6C shows the high correlation between

these species, which is an indication that LPG emissions were hom-

ogenously emitted in the whole area. Mobile sources also contrib-

ute to the levels of propane as corroborated by its correlation with

i-butane, shown in Fig. 6D, particularly with data from US, MI and

JR sites.

As envisaged, the halocarbons species showed no relationship

with compounds which are vehicular or LPG tracers. However,

some halocarbons such as vinyl chlorine and 1,2-dicholorpropane

Table 4

Concentration (ppbC) and standard deviation for the most abundant VOCs in Salamanca during FebruaryMarch 2003.

CR CG US MI JR

06:0018:00 h 18:0006:00 h 06:0018:00 h 18:0006:00 h 06:0018:00 h 18:0006:00 h 00:0023:00 h 00:0023:00 h

1 Propane 81.1 34.2 167.0 40.3 31.6 9.1 55.5 20.6 16.0 10.4 18.2 8.6 18.7 7.6 42.3 13.4

2 i-Pentane 33.9 15.4 43.6 6.7 9.3 3.3 31.8 4.4 16.0 4.6 18.0 8.0 26.9 28.7 14.0 7.7

3 n-Butane 46.2 19.7 90.8 22.9 17.9 4.4 26.7 4.9 9.2 6.0 9.0 4.2 11.4 4.4 20.5 5.3

4 Toluene 37.7 16.5 51.2 10.5 11.6 2.8 13.2 1.5 4.9 2.2 4.8 3.0 6.9 2.9 12.5 5.3

5 m/p-Xylene 28.4 13.0 37.6 15.5 7.9 2.2 8.4 2.6 5.0 1.4 8.3 8.7 22.4 36.1 10.8 7.7

6 Ethane 23.6 18.3 20.1 7.1 5.6 2.4 5.3 3.1 2.9 1.7 1.6 1.1 2.1 0.8 2.9 1.6

7 Acethylene 9.1 2.3 14.9 8.4 6.7 1.6 6.5 2.5 4.0 2.9 2.9 2.2 4.7 2.8 10.7 11.1

8 Benzene 17.2 7.1 19.5 2.5 7.3 1.7 7.2 2.8 4.2 2.5 3.3 1.9 5.0 3.4 6.9 3.2

9 i-Butane 18.0 8.4 33.1 7.5 7.7 2.3 11.1 2.9 3.4 2.2 3.3 1.9 4.2 2.5 10.8 3.2

10 Ethylene 19.4 13.6 17.2 6.5 7.3 3.6 7.6 5.3 3.3 2.4 2.4 1.8 2.5 1.1 3.4 2.2

11 Propene 16.8 10.0 18.3 8.6 3.7 1.0 4.4 0.9 1.3 0.6 1.4 0.7 0.6 0.7 2.3 3.4

12 n-Pentane 13.6 5.6 15.6 3.8 6.6 1.8 5.8 1.7 1.6 1.0 2.3 1.3 6.1 1.4 4.7 1.9

13 2,3-Dimethylbutane 10.3 6.0 14.6 2.3 3.3 1.7 4.3 0.9 2.1 0.9 2.9 1.0 2.3 1.1 5.5 3.3

14 o-Xylene 10.9 6.0 12.8 2.3 3.1 0.9 3.1 0.8 1.4 0.6 2.5 2.4 7.3 10.6 3.9 2.4

15 3-Methylbutene 41.4 71.8 18.4 4.3 0.6 0.5 0.5 0.3 5.6 3.4 0.4 0.4 5.3 9.8 1.9 0.7

16 Freon-114 8.6 2.7 14 8 a a 7.1 4.1 14.0 4.2 8.9 3.3 18.0 1.7

17 Methyl bromide a a 9.0 3.0 28.4 3.9 a 48.3 9.7 a a

18 1,2-Dichloroethane 2.8 3.0 2 0.4 9.6 3.1 31.8 4.3 0.2 0.1 49.9 10.1 1.6 1.2 2.3 1.7

19 Methyl chlorine 3.6 3.0 5.1 6.4 1.3 0.3 1.6 0.7 1.0 0.3 1.2 0.5 1.0 0.2 1.1 0.2

20 1,2-Dichloropropane 1.1 0.6 1.2 0.2 7.0 2.3 21.4 3.7 a 32.3 5.9 a a

21 Vinyl chlorine 6.1 2.7 7.7 1.5 1.9 1.8 1.7 0.6 0.3 0.2 0.4 0.3 0.6 0.2 1.5 0.8

a Not available.

0

10

20

30

0 10 20 30 40 50 60 70 80 90

MTBE(ppbC)

Toluene (ppbC)

Mobile

Industry

(A)R2=0.98

0

10

20

30

0 10 20 30 40

MTBE(ppbC)

Propene (ppbC)

Mobile

Industry

(B)

R2=0.95

0

50

100

150

200

250

300

0 30 60 90 120 150

Propane(ppbC)

n-butane (ppbC)

CR

CG

LS

MI

JR

(C)R2=0.97

0

50

100

150

200

250

300

0 10 20 30 40 50 60

Propane(ppbC)

i-butane (ppbC)

CR

CG

LS

MI

JR

(D)R2=0.92

Fig. 6. Scatter plots of selected species and its relationship with main local sources in Salamanca, Mexico. FebruaryMarch, 2003.

E. Vega et al. / Fuel 90 (2011) 35093520 3515


8/12

showed moderate association with toluene, which is probably due

to the industrial origin of the later (Table 5). On the other hand,

strong relationship was found between methyl bromide, 1,2-

dichloropropane and 1,2-dicholoroethane (Fig. 7). Methyl bromide

is mainly used as soil fumigant for the control of nematodes, fungi

and weeds; it is also used in food-processing facilities (e.g. for

extracting oils from nuts, seeds and flowers). 1,2-Dicholorporpane

is still used as pesticide, although this practice is prohibited in

North America and Europe, it has also industrial use in the applica-

tion of paint, lacquers and varnishes and for manufacture of manychlorinated compounds. 1,2-Dicholoroethane has agricultural and

industrial applications, as chemical intermediate in the vinyl chlo-

ride monomer manufacture [38].

Finally, the concentrations of isoprene, the unique biogenic

VOCs measured, showed concentrations $1.0 ppbC in the urban

sites and lower than 1.0 ppbC in the suburban and rural sites. Its

higher level in the urban area may be the result of vehicular emis-

sions rather than vegetation, which is scarce in the city.

3.3. Meteorology

Before this work, no previous studies on local wind circulation

had been reported for this region. Meteorological measurements

were carried out to determine the occurrence and properties of lo-cal flow patterns, to examine the structure and evolution of the

mixing layer over the basin and to provide data for evaluation

and testing numerical meteorological models.

Three important parameters were determined at JR, SI and VS

sites to understand transport and dilution of pollutants in this re-

gion: the mixing layer height (MH), the transport velocity (TW)

and the ventilation index (VI) which is calculated as the product

of MH and TW. In general, the VI values assured good ventilation

conditions in the whole area in the afternoon. The highest mixing

layer height was observed at 1800, reaching mean values approx-

imately of 3300 m above ground level.

The frequent meteorological conditions are exemplified with

the data from March 5: synoptic winds aloft had strong influence

on the wind flow at lowest levels, wind direction above 3 km fromthe ground is similar for the three sites; below that level, wind

direction changes slightly along the column, reaching its maximum

difference on the surface level, varying from S at JR to SSW at SI

(Fig. 8a). This meteorological condition caused the transport of pol-

lutants out of the city. On the contrary, on March 9 (Fig. 8b), wind

direction below 1 km blew from NNE at JR and VS sites to ENE at SI

site, above that level wind changed constantly until it reached syn-

optic winds (SSW), 5 km above the ground, causing transport of air

masses into the city [39].

The potential temperature evolution along the day is an impor-

tant parameter that measures the dilution of pollutants. The fre-

quent conditions observed at 0800 at all sites showed that the

surface-based inversion formed overnight was partially elimi-

nated. At noon, the mixing layer height was above 1200 m at all

sites, and at 1800 the mixing layer was approximately between

3000 and 3500 m high. On March 9, the potential temperature

had a different behavior, at 0800 at all sites presented surface-

based thermal inversion, at noon the mixing layer was below

500 m, and at 1800 the mixing layer height reached the highest

values during the field campaign, above 4000 m.

3.4. Modeling of transport and dispersion of pollutants in the region

The Regional Atmospheric Model System (RAMS) and the CIT-

SAPRC99 models were used to simulate the physical and chemical

processes which control dispersion, transport and formation of

pollutants in the atmosphere. The study region was set to an exten-

sion of 140 140 km, divided into a grid of 74 74 tridimensionalcells with a resolution of 2 2 km. the Salamanca city was located

Table 5

Correlation coefficient (Pearson) for selected halocarbon species and VOCs in

Salamanca during FebruaryMarch 2003.

MTBE i-Butane Toluene

Freon-114 0.05 0.00 0.00

Methyl bromide 0.01 0.20 0.34

1,2-Dicloroethane 0.03 0.22 0.24

Methyl chlorine 0.09 0.18 0.21

1,2-Dicloropropane 0.04 0.37 0.71Diclorobenzene 0.00 0.04 0.06

Vinyl chlorine 0.24 0.72 0.88

R2

= 0.99

R2

= 0.99

0

10

20

30

40

50

60

70

0 10 20 30 40 50 60 70

ppbC

1,2 dochloroethane (ppbC)

1,2dichloropropane

methyl bromide

Fig. 7. Halogenated compounds in ambient air in Salamanca, Mexico. FebruaryMarch, 2003.

(a) March 5, 2003: 18 hrs

Wind Direction

Altitu

de(m)

0

1000

2000

3000

4000

5000

0 60 120 180 240 300 360

VS

SI

JR

(b) March 9, 2003: 18 hrs

Wind Direction

Altitude(m)

0

1000

2000

3000

4000

5000

0 60 120 180 240 300 360

VS

SI

JR

Fig. 8. Wind direction profiles at Silao (SI), Juventino Rosas (JR) and Valle de

Santiago (VS). (a) March 5, 2003; (b) March 9, 2003 both at 18:00.

3516 E. Vega et al./ Fuel 90 (2011) 35093520


9/12

in the center of this region. The meteorological model RAMS was

used to calculate temperature, relative humidity, turbulence, wind

speed and wind direction every 300 s for each cell. The topography

and land use were taken into account for an adequate simulation of

surface wind pattern [19]. The photochemical model CIT-SAPRC99

was used to solve the transport and mass conservation of 112

chemical species using the SAPRC99 chemical mechanism [40].

Modeling results were used as complement tool to improve theunderstanding of tropospheric ozone behavior throughout the

study region where no measurements were available.

The CIT-SAPRC99 uses operator splitting technique to solve

transport and chemical processes separately in each time step. Ver-

tical diffusion is modeled as a function of atmospheric stability

class, with the mixing height entered as input to the model. The

model was modified to account for three-dimensional fields of

temperature and humidity as inputs, and to calculate reaction rates

in three dimensions on the basis of these parameters [20]. The CIT

model has been applied and tested extensively and reported else-

where [4144].

The CIT uses a terrain-following coordinate system and 15 ver-

tical layers were defined for this work, with the top boundary of

the modeling domain at 4628 m above the surface. The height of

each layer became smaller as it approached to the surface (e.g.

50 m for the lowest layer) for greater resolution. The wind fields

obtained from RAMS model were interpolated to fit the CIT-SAP-

RC99 grid domain, smoothed and filtered to improve mass consis-

tency in the CIT-SAPRC99 model.

The results obtained from the RAMS model indicated two differ-

ent dispersion conditions during the study period. The first and

more frequent condition showed circulation of winds during early

morning and night that converged towards the Salamanca city i.e.mountain-valley circulation (Fig. 9A). From 1000, the circulation

shifted towards the mountains (Fig. 9B) with a NW wind direction

at 1800 (Fig. 9C). After this time, the wind speed decreased and the

wind direction returned towards the city (Fig. 9D). The second less

frequent condition indicated a constant wind direction from the

NE, from 1000 to 2000 (not shown). Such meteorological condition

led to accumulation of emissions inside the valley and therefore

the increase of pollution levels.

The results of coupled application of RAMS and CIT-SAPRC99

models to estimate production and transport of ozone are illus-

trated in Fig. 10A-D, which describe ozone concentration indicated

with background colored from blue (0 ppb) to red (200 ppb), wind

fields (black arrows) and VOCs plumes (green zones) emitted from

Len, Irapuato, Salamanca and Celaya cities (L, I, S and C yellow let-

ters, respectively). Simulations were calculated for 0900, 1200,

(A) 0800. (B) 1000.

(C) 1800. (D) 2300.

Fig. 9. Surface wind field determined by RAMS model for the study region on March 4, 2003 at (A) 0800; (B) 1000; (C) 1800 and (D) 2300.

E. Vega et al. / Fuel 90 (2011) 35093520 3517


10/12

1600 and 2100. Fig. 10A shows that wind flow at 0900 was domi-

nated by drainage flow of cold air descending from the mountains,

transporting urban emissions to the lowest areas of the region. Athis time of the day, high atmospheric stability conditions and ther-

mal inversion caused that emissions were trapped near the surface

and minimum ozone production. At noon, VOCs were diluted due

to the rapid growth of the mixing height, so only a small plume

is observed at L. Ozone production is evident in those regions

where precursors were transported to early in the morning, while

ozone concentrations were minimal in the urban areas. Fig. 10C

shows that wind pattern at 1600 has changed, transporting air

masses from southwest to northeast. The ozone formation was

well developed in the whole region and the maximum ozone con-

centrations were calculated in the NE of the models domain.

Again, at night (Fig. 10D), cold air descending from the mountains

began to dominate the wind circulation in the region. As expected,

the CIT-SAPRC99 predicted moderate ozone concentration (4060 ppb) remaining at surface level in the NE rural zones due to

insignificant of NOX sources in that region and again VOCs concen-

trations became significant at surface level in the urban zones.

The above described behavior was frequently observed in theregion, especially when weak synoptic conditions generated val-

ley-mountain circulation. Nevertheless, it is not possible to deter-

mine a general wind pattern due to the limited time of the

study; it would be necessary to carry out the analysis for other sea-

son of the year.

Finally, overall observed and modeled results of this survey

indicate environmental implications due to the exposition of pop-

ulation to secondary pollutants and halogenated hydrocarbons,

especially in the rural areas located northeast of the industrial

and urban zones. As a consequence of this study, local and federal

environmental authorities in collaboration with PEMEX (the Na-

tional Oil Company) and the Comisin Federal de Electricidad (Fed-

eral Electricity Bureau) developed the first Air Quality Program for

the region (20032006). Main achievements of this program werethe reduction in number of days with sulfur dioxide concentrations

(A) 0900. (B) 1200.

(C) 1600. (D) 2100.

L

I

S

C

Fig. 10. Hydrocarbon urban plumes (green) and ozone concentrations (from blue = 0 ppb to red = 200 ppb) modeled by coupled application of RAMS and CIT-SAPRC99 during

March 4 2003 in Salamanca region. (A) 0900; (B) 1200; (C) 1600 and (D) 2100. Arrows indicate surface winds. Panel A shows the location of cities: L = Leon; I = Irapuato;

S = Salamanca and C = Celaya. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3518 E. Vega et al./ Fuel 90 (2011) 35093520


11/12

above the standard, from 79 in 2003 to 34 in 2006. Later on, the

program included definitive removal of methyl bromide usage for

agricultural activities [24].

4. Conclusions

A 2-week field campaign of measurements of criteria pollutants,VOC and meteorological parameters was carried out during Febru-

aryMarch 2003 in the Salamanca city which is located inside one

of the most important industrial corridors of Mexico, to characterize

airqualityin theurban andregional scales. Special emphasiswas gi-

ven to the chemical characterization of VOC to gain knowledge

about levels, distribution and origin of these species. In spite of lim-

ited period of time, this monitoring effort currently represents one

of the biggest studies on airqualityrealized in Mexico, without con-

sideringthe mostrecent studies carried out in the metropolitan area

of Mexico City, which allows investigating the air quality condition

in an area highly influenced by industrial emissions.

Regarding one of the most important pollutants, the SO2showed a high concentration event at 0200 at the urban site, lo-

cated 2 km west of the Refinery and the Power Plant. The levels

of this pollutant reached 0.60 ppm (1 min sampling-time). Histor-

ically, extremely high levels of SO2 have been frequently observed

from 0000 to 0300 in the city, attributed to venting operations of

the Refinery and the Power Plant. However since 2003 control

measurements have been implemented to reduce levels of SO2and particles in the atmosphere, such as the shifting to cleaner

fuels (with lower content of S) in both the Power Plant and the

Refinery. These measurements were already in execution by the

time of the field campaign, thus it is possible that the first positive

results were observed.

The highest ozone levels were observed in the rural monitoring

sites, which are explained by transport and possible inputs from

the free troposphere, given the high altitude of the study region.

Total VOCs mass concentration fluctuated from 170 ppbC in the

rural area to 956 ppbC in the urban sites. The more abundant groupwas the alkanes followed by aromatics, olefins and halogenated.

The total mass concentration of VOCs is comparable with

values reported for some urban and rural sites of Mexico City

Metropolitan Area. However, the concentration of halogenated

species (particularly Freon-114, methyl bromide, 1,2-dichloroeth-

ane and 1,2-dichloropropane) was notoriously higher in rural sites

of Salamanca compared with Mexico City and other highly polluted

areas of China. Formaldehyde and acetaldehyde showed average

concentrations of 4.3 and 2.7 ppb, respectively which are lower

than those reported for Mexico City (13.3 and 4.4 ppb respec-

tively). Due to these low values, it is expected little contribution

of carbonyl species to the photochemical activity in the region.

The origin of VOCs was inferred by examining the concentration

and spatial distribution of species, in conjunction with cross corre-lation analyses and relationship of marker species for specific

sources. In the urban area, gasoline and diesel vehicular exhaust

and marketing/leaking of LPG dominated the VOCs burden, while

emissions from industrial and agriculture activities were impor-

tant in the rural area. The last two sources emit halogenated spe-

cies that represent potential hazard to human health due to the

toxic and/or carcinogenic effect, thus the results of this study high-

lighted the needed of monitoring and even implementing control

measurements for these species. As a consequence, federal and lo-

cal environmental authorities removed usage of methyl bromide in

agricultural activities since 2007.

Meteorological conditions generally favoured the transport of

pollutants outside Salamanca towards northeast, where the high-

est concentration of ozone were observed. This result was in agree-ment with the modeling calculations.

Acknowledgements

This study was supported by Mexico national oil company

(PEMEX) under contract IMP-ZC-001-2003. The authors are grate-

ful to the authorities of Salamanca refinery by their help and sup-

port provided during the study.

References

[1] Shahriar S, Topal E. When will fossil fuel reserves be diminished? Energy Policy2009;37:1819.

[2] Interngovermental Panel on Climate Change (IPCC). Climate change 2001: thescientific basis. Contribution of the working group I to the third assessmentreport of the interngovermental panel on climate change. Technicalsummary. Cambridge International Press; 2001.

[3] Decker EH, Elliot S, Smith FA, Blake DR, Rowland FS. Energy and material flowthrough the urban environment. Ann Rev Energy Environ 2000;25:685740.

[4] Molina LT, Kolb CE, de Foy B, Lamb BK, Brune WH, Jimenez JL. Air quality inNorth Americas most populous city overview of the MCMA-2003 campaign.Atmos Chem Phys 2007;7:244773.

[5] Air Resources Board of California, USA. [cited 22.01.11].

[6] Secretara de Energa. Prospectiva de petrolferos 20062015. ISBN: 968-874-201-5 DF, Mexico; 2006.

[7] Instituto Nacional de Estadtica Geografa e Informtica. XII Censo General dePoblacin y Vivienda 2000. DF, Mxico; 2001.

[8] SEMARNAT-IEG. Programa para mejorar la calidad del aire en la ciudad deSalamanca (PROAIRE). Peridico Oficial del Estado de Guanajuato No. 9, 6

junio: Mxico; 2003.[9] Seinfeld JH, Pandis SN. Atmospheric chemistry and physics: from air pollution

to global change. New York (NY): John Wiley and Sons; 1998.[10] Jacobson MZ. Atmospheric pollution, history, science, and regulation. United

Kingdom: Cambridge University Press; 2002.[11] Jenkin ME, Clemitshaw KC. Ozone and other secondary photochemical

pollutants: chemical processes governing their formation in the planetaryboundary layer. Atmos Environ 2000;34:2499527.

[12] Atkinson R. Atmospheric chemistry of VOCs and NOX. Atmos Environ2000;34:2063101.

[13] Atkinson R, Arey J. Gas phase tropospheric chemistry of biogenic organiccompounds: a review. Atmos Environ 2003;37:S197119.

[14] Odum JR, Hoffmann T, Bowman F, Collins D, Flagan RC, Seinfeld JH. Gas/particle partitioning and secondary organic aerosol yields. Environ Sci Technol1996;30:25805.

[15] Molina MJ, Rowland FS. Stratospheric sink for chlorofluoromethanes:chlororine atom catalysed destruction of ozone. Nature 1974;249:8102.

[16] Vega E, Ruiz H, Martinez-Villa G, Sosa G, Gonzalez-Avalos E, Reyes E, et al. Fineand coarse particulate matter characterization in a heavily industrialized cityin central Mexico during Winter 2003. J Air Waste Manage Assoc2007;57:62033.

[17] Environmental Protection Agency. Compendium method TO-14a.Determination of volatile organic compounds (VOCs) in ambient air usingspecially prepared canisters with subsequent analysis by gas chromatography(EPA/625/R-96/010b). Compendium of methods for the determination of toxicorganic compounds in ambient air. 2nd ed. Cincinnati (OH); 1999.

[18] McCleanny WA, Plei JD, Evans GF. Canister based method for monitoring toxicVOCs in ambient air. J Air Waste Manage Assoc 1991;41:130818.

[19] Pielke RA. A comprehensive meteorological model system: RAMS. MeteorolAtmos Phys 1992;49:6991.

[20] West JJ, Zavala MA, Molina LT, Molina MJ, San Martini F, McRae GJ, et al.Modeling ozone photochemistry and evaluation of hydrocarbon emissions inthe Mexico City metropolitan area. J Geophys Res 2004. doi:10.1029/

2004JD004614.[21] Vingarzan R. A review of surface ozone background levels and trends. AtmosEnviron 2004;38:343142.

[22] Kalabokas PD, Repapis CC. A climatological study of rural surface ozone inCentral Greece. Atmos Chem Phys 2004;4:113947.

[23] Kalabokas PD, Volz-Thomas A, Brioude J, Thouret V, Cammas JP, Repapis CC.Vertical ozone measurements in the troposphere over the EasternMediterranean and comparison with Central Europe. Atmos Chem Phys2007;7:378390.

[24] SEMARNAT/Gobierno del Estado de Guanajuato. Progama para mejorar lacalidad del aire en Salamanca 20072012. Mexico; 2007.

[25] Velasco E, Lamb B, Westberg H, Allwine1 E, Sosa G, Arriaga-Colina JL, et al.Distribution, magnitudes, reactivities, ratios and diurnal patterns of volatileorganic compounds in the Valley of Mexico during the MCMA 2002 & 2003field campaigns. Atmos Chem Phys 2007;7:32953.

[26] Chan CY, Tang JH, Li YS, Chan LY. Mixing ratios and sources of halocarbons inurban, semi-urban and rural sites of the Pearl River Delta, South China. AtmosEnviron 2006;40:733145.

[27] Barletta B, Meinardi S, Simpson IJ, Rowland FS, Chan C, Wang X, et al. Ambient

halocarbon mixing ratios in 45 Chinese cities. Atmos Environ2006;40:770619.

E. Vega et al. / Fuel 90 (2011) 35093520 3519
http://www.arb.ca.gov/toxics/id/taclist.htmhttp://www.arb.ca.gov/toxics/id/taclist.htmhttp://dx.doi.org/10.1029/2004JD004614http://dx.doi.org/10.1029/2004JD004614http://dx.doi.org/10.1029/2004JD004614http://dx.doi.org/10.1029/2004JD004614http://www.arb.ca.gov/toxics/id/taclist.htmhttp://www.arb.ca.gov/toxics/id/taclist.htm


12/12

[28] Lei W, de Foy B, Zavala M, Volkamer R, Molina LT. Characterizing ozoneproduction in the Mexico City Metropolitan Area: a case study using achemical transport model. Atmos Chem Phys 2007;7:134766.

[29] Grutter M, Flores E, Andraca-Ayala G, Baez A. Formaldehyde levels indowntown Mexico City during 2003. Atmos Environ 2005;39:102734.

[30] MIT-CAM. The MCMA Field Campaign; 2003. [cited 2010 August].

[31] Vega E, Mugica V, Carmona R, Valencia E. Hydrocarbon source apportionmentin Mexico City using the chemical mass balance receptor model. AtmosEnviron 2000;34:41219.

[32] Blake DR, Rowland FS. Urban leakage of liquefied petroleum gas and its impacton Mexico City Air Quality. Science 1995;269:9536;Watson JG, Chow JC, Fujita EM. Review of volatile organic compounds sourceapportionment by chemical mass balance. Atmos Environ 2001;35:156784.

[33] Fujita EM. Hydrocarbon source apportionment for the 1996 Paso del NorteOzone Study. Sci Total Environ 2001;276:17184.

[34] Environmental Protection Agency [Internet]. USA: SPECIATE4.2 data browser. [updated 2008November; cited 2001 January 21].

[35] CoxML,Fraser PJ,Sturrock GA,Siems ST,PorterLW. Terrestrial sources andsinksof halomethanes near Cape Grim, Tasmania. Atmos Environ 2004;38:383952.

[36] Crounse JD, DeCarlo PF, Blake DR, Emmons LK, Campos TL, Apel EC, et al.Biomass burning and urban air pollution over the Central Mexican Plateau.Atmos Chem Phys 2009;9:492944.

[37] de Gouw JA, Welsh-Bon D, Warneke C, Kuster WC, Alexander L, Baker AK, et al.During the MILAGRO study. Atmos Chem Phys 2006;2009:9.34253442.

[38] International Program on Chemical Safety [Internet] Chemical SafetyInformation from Intergovernmental Organizations. [Updated 21.09.10; cited 21.01.11].

[39] Instituto Mexicano del Petrleo. Estudio local y regional del impacto de lasemisiones de PEMEX y CFE en la calidad del aire en Salamanca, Gto. FinalReport. PS-MAS-IF-F21026-02. DF, Mxico; 2003.

[40] Carter WPL. Documentation of the SAPRC-99 chemical mechanism for VOCreactivity, final report to California Air Resources Board, Contract 92-329 and

95-308, Calif. Air Res. Board, Sacramento, Calif. USA. [cited 19.03.11].

[41] McRae GJ, Goodin WR, Seinfeld JH. Development of a second-generationmathematical model for urban air pollution: 1. Model formulation. AtmosEnviron 1982;16:67996.

[42] Los Alamos National Laboratory and Instituto Mexicano del Petroleo. Mexicocity air quality research initiative (MARI), rep LA-12699, Los Alamos, N.M.;1994.

[43] Harley RA, Sawyer RF, Milford JB. Updated photochemical modeling forCalifornias south coast air basin: comparison of chemical mechanisms andmotor vehicle emission inventories. Environ Sci Technol 1997;31:282939.

[44] Marr LC, Noblet GS, Harley RA. Formation of photochemical air pollution incentral California: 2. Impact of revised emissions on Eulerian modelpredictions. Geophys Res 2002;107(D6):4048. doi:10.1029/2001JD00069.

3520 E. Vega et al./ Fuel 90 (2011) 35093520
http://mce2.org/fc03/fc03.htmlhttp://mce2.org/fc03/fc03.htmlhttp://www.epa.gov/ttn/chief/software/speciate/index.htmlhttp://www.inchem.org/http://www.cert.ucr.edu/~carterhttp://www.cert.ucr.edu/~carterhttp://dx.doi.org/10.1029/2001JD00069http://dx.doi.org/10.1029/2001JD00069http://www.cert.ucr.edu/~carterhttp://www.cert.ucr.edu/~carterhttp://www.inchem.org/http://www.epa.gov/ttn/chief/software/speciate/index.htmlhttp://mce2.org/fc03/fc03.htmlhttp://mce2.org/fc03/fc03.html

art salamanca fuel 90 2011

Documents