This article was downloaded by: [201.170.123.

1]
On: 21 November 2014, At: 13:49
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of the Air & Waste Management

Association
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/uawm20

Trace Gases and Particulate Matter Emissions from

Wildfires and Agricultural Burning in Northeastern
Mexico during the 2000 Fire Season
a

Alberto Mendoza , Marisa R. Garcia , Patricia Vela , D. Fabian Lozano & David
Allen

Department of Chemical Engineering , Instituto Tecnolgico y de Estudios

Superiores de Monterrey , Monterrey , Mexico
b

Center for Environmental Quality, Instituto Tecnolgico y de Estudios Superiores de

Monterrey , Monterrey , Mexico
c

Department of Chemical Engineering, The University of Texas at Austin , Austin ,

TX , USA
Published online: 01 Mar 2012.

To cite this article: Alberto Mendoza , Marisa R. Garcia , Patricia Vela , D. Fabian Lozano & David Allen (2005)
Trace Gases and Particulate Matter Emissions from Wildfires and Agricultural Burning in Northeastern Mexico
during the 2000 Fire Season, Journal of the Air & Waste Management Association, 55:12, 1797-1808, DOI:
10.1080/10473289.2005.10464778
To link to this article: http://dx.doi.org/10.1080/10473289.2005.10464778

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the Content)
contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors
make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any
purpose of the Content. Any opinions and views expressed in this publication are the opinions and views
of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources of information.
Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs,
expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in
connection with, in relation to or arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

TECHNICAL PAPER

ISSN 1047-3289 J. Air & Waste Manage. Assoc. 55:17971808

Copyright 2005 Air & Waste Management Association

Trace Gases and Particulate Matter Emissions from Wildfires

and Agricultural Burning in Northeastern Mexico during the
2000 Fire Season
Alberto Mendoza
Department of Chemical Engineering, Instituto Tecnologico y de Estudios Superiores
de Monterrey, Monterrey, Mexico

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Marisa R. Garcia, Patricia Vela, and D. Fabian Lozano

Center for Environmental Quality, Instituto Tecnologico y de Estudios Superiores de Monterrey,
Monterrey, Mexico
David Allen
Department of Chemical Engineering, The University of Texas at Austin, Austin, TX

ABSTRACT
An inventory of air pollutants emitted from forest and
agricultural fires in Northeastern Mexico for the period of
January to August of 2000 is presented. The emissions
estimates were calculated using an emissions factor methodology. The inventory accounts for the emission of carbon monoxide (CO), methane, nonmethane hydrocarbons, ammonia, nitrogen oxides, and particulate matter
(PM). Particulate matter emissions include estimates for
fine PM and coarse PM. A total of 2479 wildfires were
identified in the domain for the period of interest, which
represented 810,000 acres burned and 621,130 short
tons emitted (81% being CO). The main source of information used to locate and estimate the extent of the fires
came from satellite imagery. A geographic information
system was used to determine the type of vegetation
burned by each fire. More than 54% of the total area
burned during the period of study was land on the State of

IMPLICATIONS
Contribution of wildfires to the total emissions of air pollutants in Mexico is not well understood, nor is the potential
impact of those emissions in the air quality of the country
and abroad. We present a study that will help future analyses to apportion and model the contribution of wildfires to
air quality degradation in northern Mexico and southern
United States. Furthermore, future regional air quality observational programs can benefit in their planning process
from the results presented here as periods and areas with
greater potential to ignite are identified, and, thus, the observational program can try to capture the effects of such
fire events.

Volume 55 December 2005

Tamaulipas. However, 58% of the estimated emissions

came from the State of Coahuila. This was because of the
mix of vegetation types burned in each state. With respect
to the temporal distribution, 76.9% of the fires occurred
during the months of April and May consuming almost
78% of the total area burned during the period of study.
Analysis of wind forward trajectories of air masses passing
through the burned areas and 850-mb wind reanalyses
indicate possible transboundary transport of the emissions from Mexico to the United States during the occurrence of the major wildfires identified.
INTRODUCTION
Emissions from wildfires and prescribed agricultural burning can have an important contribution in the degradation of air quality at the local, regional, and global scales.
Impacts from the extensive occurrence of vegetation fires
include an increase in carbon dioxide (CO2) emissions
and the corresponding effect on global warming;13 increase in particulate matter (PM) levels, particularly of the
fine fraction, with the subsequent degradation of visibility;4 6 and overall changes in the tropospheric chemistry
of trace gases.710 Studies also suggested that wildfires
could contribute as an important source of toxics, like
mercury,11 and stratospheric ozone-depleting compounds, like methyl bromide.12
Methods that have been used to determine the contribution of wildfires to air pollution include the use of
satellite imagery to examine the spatial extent of smoke
plumes,13 direct measurement of trace gases14 and PM
(including chemical markers)15 in ambient air, and source
Journal of the Air & Waste Management Association 1797

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Mendoza, Garcia, Vela, Lozano, and Allen

measurements16,17 for the development of emission factors and speciation profiles that can then
be used in source apportionment and modeling
studies.
Past studies18 20 have documented the effect
that extreme wildfire events occurring in southern Mexico and Central America can have on the
air quality in areas of the United States. However,
few efforts have been conducted to quantify
emissions from particular Mexican regions
(other than Mexico City21) derived from wildfires
during nonextreme events. Here we use a methodology based on emissions factors to estimate
the emission of trace gases and PM from forest,
grasslands, rangelands, and agricultural fires in
northeast Mexico from January through August
of 2000. The inventory was constructed with the
major compounds released to the atmosphere
during an outdoor fire: carbon monoxide (CO),
methane (CH4), nonmethane hydrocarbons, ammonia (NH3), nitrogen oxides (NOx), and PM.
PM was categorized into coarse PM (PM10) and
fine PM (PM2.5). The motivation of the study is to Figure 2. Emission factors methodology followed to estimate the emissions
obtain additional information that can be used inventory.
in ongoing efforts22 to better characterize the
months of April and May. Of the three Mexican States
possible contribution of Mexican emissions in the air
quality of Texas.
that comprise the region (Coahuila, Nuevo Leon, and
Tamaulipas), the first two have very similar vegetation,
Description of the Region of Study
topography, and climate. Tamaulipas is different, because
Northeastern Mexico (see Figure 1) is a region that is
it presents also coastal and subtropical conditions. Northcharacterized by having 70% of its wildfires during the
east Mexico is, in general, a dry and hot region.
The Sierra Madre Oriental is the main physiographic
feature that crosses all three of the states. The mountain
range begins in the south near the city of Pachuca in the
state of Hidalgo and runs parallel to the gulf coast all along
the southern half of Tamaulipas, entering Nuevo Leon,
where, in the proximity of Monterrey, it changes to an
east-west direction; then crosses to Coahuila until it reaches
the U.S.Mexico border close to the area of the Big Bend
National Park (on the U.S. side). The Sierra Madre creates a
clear division between the lowlands of Tamaulipas and the
uplands of western Nuevo Leon and the entire state of
Coahuila. Thus, the Sierra Madre defines the following main
regions of the study area: (1) the coastal area of Tamaulipas
with a subtropical climate; (2) the Sierra Madre itself, a
mountainous system with elevations that reach up to

Figure 1. Political limits of Northeastern Mexico.

1798 Journal of the Air & Waste Management Association

3000 m above sea level, and relatively temperate climate;

and (3) the hot and dry interior lands in Nuevo Leon and
Coahuila.
The dry climate of Coahuila and Nuevo Leon limits
the agricultural activities to those areas where irrigation
systems can provide water for the crops. However, many
Volume 55 December 2005

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Mendoza, Garcia, Vela, Lozano, and Allen

areas of both states (and some portions in northwesternTamaulipas) are dedicated to cattle ranching. The extraction of forest products is limited by both the relatively
small area occupied by forest and the difficulties in reaching those areas in the Sierra Madre.
Tamaulipas is a more diverse state. It has important
mountains, coastland, scrubs, tropical areas, and forests.
Tamaulipas weather is influenced by mountains that run
through it in a north to south direction and also by the
distance to the Gulf of Mexico. Tropical and humid regions of Tamaulipas are located in the south. Coast plains
and almost all of the other plains are dry, with scrub
vegetation.
During the spring, northeast Mexico is characterized
by very dry conditions; between the months of January
and April, Nuevo Leon and Coahuila receive 180 mm of
rain, and Tamaulipas receives even less. April is a key
month: after 5 months of dry conditions, the temperature
increases dramatically. This sudden change in temperature makes the land suitable for fire ignition and spreading. The rain season for the entire region begins in late
August and September. For example, August is a hot
month with an average temperature similar to April, but
with much more rain (e.g., 100 mm in Nuevo Leon).
Temperatures decrease 23 C on average during the second half of the summer. This increase in precipitation, in
combination with the temperature drop, makes it very
hard for a fire to begin and spread during July and August,
as fire records show.

a potential fire pixel. Cloud cover is usually a significant

factor in identification of fires from satellite data. In this
study, the technique described by Saunders and Kriebel25
was used to screen out the cloud cover in the imagery
used. All of the pixels of interest were classified using a
geographic information system. The fire coverage was
then overlapped onto vegetation coverage to obtain the
vegetation types affected by each fire. The primary source
of the vegetation maps used was the CONABIO correction
to the Mexico National Statistics, Geography, and Informatics Institute (INEGI) 1:250,000 map series for the period 19811991.26
Additional efforts to obtain information related to
wildfires, including their location, duration, and acreage

Development of the Inventory

Estimation of Area and Vegetation Classes Burned. The procedures used were based, in part, on previous work conducted to derive a similar emissions inventory, but for the
state of Texas.2224 The bulk of the fire database was created using satellite data obtained from National Commission for the Understanding and Use of Biodiversity
(CONABIO). A hierarchical procedure was used to analyze
the advanced very high-resolution radiometer data sets
that were obtained from CONABIO through its HeatPoints (Fires) Monitoring Program. Satellite images from
January to August of 2000 were processed, and spatial
coordinates for each pixel catalogued as fire were assigned. Data acquisition times for the satellite images used
were 10 hr (Greenwich Mean Time [GMT]) and 21 or 22 hr
(GMT). Any fire that occurred between these acquisitions
and was of short duration was not detected. One of the
assumptions is that such short-duration fires may have
very little effect in the amount of emission accumulated
for the period of this study. In addition, to detect a heat
point it must have a minimum area of 120 m2 (0.1% of
the pixel size) and a temperature 250 C. This creates a
pixel with a nominal temperature of 50 C and, therefore,

Mixed desert scrub

Volume 55 December 2005

Table 1.
types.

Mapping of Mexican vegetation classes to FOFEM 5.0 cover

Mexican Vegetation Class

Oak
SW shrub steppe
Riparian woodland
Pine
Evergreen forest
Evergreen seasonal forest
Semideciduous seasonal forest
Semievergreen seasonal forest
Chaparral

Desert scrub
Halophytic scrub
Desert grassland
Gypsophilous scrub
Gypsum grassland
Cultivated grassland
Grassland
Grassland-acacia
Aquatic vegetation
Mosaic cropland
Deciduous seasonal forest
Tamaulipan thornscrub
Thorn forest
Sand dune vegetation
Secondary grassland
Mesquite shrub
Juniper
Palm

FOFEM Vegetation Class

SAF 42 Bur Oak
SAF 67 Mohrs (shin) Oak (moderate
cover)
SAF 95 Black Willow
SRM 109 Ponderosa Pine Shrublands
SRM 202 Coast Live Oak Woodland
SRM 205 Coastal Sage Shrub
SRM 207 Scrub Oak Mixed Chaparral
SRM 207 Scrub Oak Mixed Chaparral
SRM 208 Ceanothus Mixed Chaparral
SRM 211 Creasote Bush Scrub (moderate
shrub cover)
SRM 211 Creasote Bush Scrub (moderate
shrub cover)
SRM 401 Basin Big Sagebrush - (moderate
shrub cover)
SRM 701 Alkali SacatonTobosagrass
SRM 701 Alkali SacatonTobosagrass
SRM 701 Alkali SacatonTobosagrass
SRM 702 Black GramaAlkali Sacaton
SRM 702 Black GramaAlkali Sacaton
SRM 702 Black GramaAlkali Sacaton
SRM 704 Blue GramaWestern Wheatgrass
SRM 704 Blue GramaWestern Wheatgrass
SRM 719 MesquiteLiveoak - Seacoast
Bluestem
SRM 719 MesquiteLiveoak - Seacoast
Bluestem
SRM 719 MesquiteLiveoak - Seacoast
Bluestem
SRM 722 Sand SagebrushMixed Prairie
SRM 722 Sand SagebrushMixed Prairie
SRM 727 MesquiteBuffalograss
SRM 733 JuniperOak
SRM 814 Cabbage Palms Flatwood

Journal of the Air & Waste Management Association 1799

Mendoza, Garcia, Vela, Lozano, and Allen

Table 2. Emission factors (lb of pollutant/t of fuel burned) estimated for
each fuel component.
Fuel Component
Litter, wood 01 in.
Wood 13 in.
Wood 3 in.
Herb, shrub, regen
Duff
Canopy fuels

CO

CH4

NMHC

PM2.5

PM10

NOx

NH3

52.4
111.4
170.5
249.2
308.2
249.2

3.3
5.9
8.5
12.0
14.6
12.0

5.6
8.8
12.00
16.3
19.4
16.3

7.9
11.9
15.9
21.2
25.2
21.2

9.3
14.0
18.7
25.1
29.8
25.1

2.5
2.5
2.5
2.5
2.5
2.5

0.5
1.1
1.7
2.6
3.2
2.6

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Notes: CH4 methane; NMHC nonmethane hydrocarbon.

burned, included communications with local, state, and

federal agencies. CONAFOR (Mexican forestry agency)
and Proteccion Civil (Mexican emergency service) provided data for fire events occurring during the months of
April to September of 2000. The database created using
satellite images was complemented with this information.
However, it needs to be noted that the number of fires and
area burned reported by CONAFOR and Proteccion Civil
was considerably smaller than the obtained through the
analysis of satellite imagery. For example, during August
and September, CONAFOR reported seven wildfires for
the three states that consumed a total of 671 acres; the
largest fire of the seven reported consumed 370 acres. In
comparison, the total estimated area burned for August
(imagery-derived plus the additional sources) gave a total
of 28,590 acres.
According to the Ministry of the Environment and
Natural Resources and the Ministry of Agriculture, Rural
Development, Fisheries, and Food, prescribed burning in
rangelands or agricultural lands is not practiced in Mexican northeastern states. The information provided by the
agencies suggests no important contribution of this practice to the total emissions. An expert on agriculture and
cattle breeding practices in Mexico indicated that, indeed,
prescribed burning is a method no longer used in Northern Mexico to improve or clear the soil; this is, in part,
because of an intense government campaign to educate
Table 3.

farmers.27 Agricultural and cattle breeding associations

were contacted in 12 municipalities in the three states
in search of additional information. Each of the 12 associations contacted confirmed the information previously
given: no prescribed burns are conducted. In particular,
the Cattle Breeding Association of Montemorelos (a municipality of the state of Nuevo Leon) indicated that there
are some burns of crop residues, especially during the
harvest of sorghum, but these burns are conducted during
October and November. During the period of interest,
they reported no controlled fires, except for road clearing
(no official statistics on this activity are kept). Sugar cane
prescribed burning does take place, in particular in the
state of Tamaulipas. However, no detailed information on
location of these prescribed burns was available. Based on
the above information, the inventory presented here does
not take into account explicit emissions from prescribed
agricultural burning. All of the fires identified through
satellite imagery that occurred in areas covered by agricultural vegetation were considered wildfires. No direct
identification of the type of crop being burned was attempted, because the emissions factor model used does
not make any differentiation between subcategories of
croplands or cultivated grasslands.
Emission Estimation Procedures. The algorithm for the
emissions estimation use in this work is shown in Figure
2. As mentioned previously, a procedure based on computing emission factors was applied.
First, consumed loads or fuel loads (i.e., mass of fuel
consumed per unit area), CLik, for each fuel type i and
vegetation class k burned were estimated using the First
Order Fire Effects Model (FOFEM) version 5.28 The selection of FOFEM as the modeling tool was based on the
need to have a consistent modeling methodology across
source categories and to have an inventory that could be
compared with previous work developed for the state of
Texas.22 The fuels categories used by FOFEM are litter,
duff, dead woody fuel (wood 0 1 in., wood 13 in., and

Composite emission factors calculated for whitebark pine cover type for spring conditions.

Fuel
Component

Consumed
Load (t/acre)

Litter
Wood 13 in.
Wood 3 in.
Herbaceous
Shrub
Total

0.30
0.50
1.88
0.20
0.24

Composite Emission Factors (lb/acre)

CO

CH4

NMHC

PM2.5

PM10

NOx

NH3

15.720
55.720
320.502
49.840
73.978
515.760

0.996
2.956
15.988
2.392
3.492
25.824

1.683
4.402
22.554
3.251
4.668
36.558

2.364
5.944
29.885
4.248
6.059
48.500

2.789
7.013
35.263
5.013
7.150
57.230

0.738
1.230
4.625
0.492
0.590
7.675

0.153
0.551
3.214
0.510
0.769
5.198

Notes: CH4 methane; NMHC nonmethane hydrocarbon.

1800 Journal of the Air & Waste Management Association

Volume 55 December 2005

Mendoza, Garcia, Vela, Lozano, and Allen

Table 4. Summary of number of fires, total burned area, and air
pollutant emissions estimated for northeastern Mexico between January
and August 2000.
State

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Variable

Coahuila

Nuevo
Leon Tamaulipas

No. of fires
737
427
Total burned area (acres)
252,740 115,263
Total emissions (thousands short t)
CO
296
54.9
CH4
14.0
2.61
NMHC
18.8
3.51
PM2.5
24.4
4.57
PM10
28.8
5.39
NOx
2.44
0.48
NH3
3.08
0.57

Total

1,315
441,568

2,479
809,570

154
7.35
9.90
12.9
15.2
1.37
1.60

506
24.0
32.2
41.9
49.4
4.29
5.24

Notes: CH4 methane; NMHC nonmethane hydrocarbon.

wood 3 in.), and live fuel (herb, shrub, conifer regeneration or regen, and canopy fuel components). Consumption of woody fuels is based on the BURNUP model,29,30
which is a physical model of heat transfer and burning
rate of woody fuel particles as they interact over the
duration of a burn. In addition, separate estimates of
flaming and smoldering consumption are also available.
Consumption of other fuels is predicted using a variety of
empirical equations and rules of thumb.28 One major

assumption that FOFEM makes in predicting fuel consumption is that the entire area of concern experienced
fire.
To compute the consumed loads, FOFEM 5 requires
as input the vegetation classes involved in each wildfire.
Mexican vegetation types are classified by INEGI, whereas
the vegetation types used with FOFEM 5 were the ones
specified by the Society of American Foresters for forested
areas and the Forest-Range Environmental Study ecosystem types for shrub and grassland areas. In addition,
FOFEM cover types are grouped into four geographic regions. Thus, the Mexican vegetation types defined by
INEGI were mapped to FOFEM cover types based on previous studies reported in the literature.3135 The cover
types available in FOFEM regions Interior West and
South East were the ones selected for the mapping procedure. The final mapping of vegetation types used for
this project is shown in Table 1. It is important to note
that the vegetation classes reported in Table 1 are those
exclusively found in the area of interest. Once the vegetation types were identified, the FOFEM fuel consumption
model was applied. Parameters set in the model application include very dry moisture conditions at time of
burn, natural fuel as the fuel condition, typical fuel
loadings, default log rotten percentage, and even log
loading distribution. The results obtained were consumed
loads of the fuels cited earlier, for each vegetation type, as
a function of the season of the burn.

Figure 3. Area affected by wildfires for each vegetation class considered in the study for the period of January to August 2000.
Volume 55 December 2005

Journal of the Air & Waste Management Association 1801

Mendoza, Garcia, Vela, Lozano, and Allen

Table 5. Composite emission factors (lb/acre)a estimated for the vegetation types with the most occurrences of wildfires.
Vegetation Type
(Mexican Classification)
Shrub steppe
Chaparral
Mixed desert scrub, desert scrub
Desert grassland, gypsophilous scrub,
gypsum grassland
Cultivated grassland, grassland,
grassland-acacia
Aquatic vegetation, mosaic cropland
Deciduous seasonal forest, tamaulipan
thornscrub, thorn forest

CO

CH4

NMHC

PM2.5

PM10

NOx

NH3

527.7
343.1
616.6

25.09
17.19
29.24

33.76
24.36
39.26

43.95
32.33
51.07

51.86
38.15
60.26

4.625
5.141
5.240

5.465
3.467
6.396

142.0

6.817

9.265

12.11

14.29

1.402

1.454

142.0
142.0

6.817
6.817

9.265
9.265

12.11
12.11

14.29
14.29

1.402
1.402

1.454
1.454

1493

70.83

95.17

123.8

146.1

12.79

15.46

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Notes: CH4 methane; NMHC nonmethane hydrocarbon; aSelected CEFs are for spring season; CEFs for other seasons follow the same general pattern shown
here.

Second, emission factors (mass of pollutant emitted

per unit mass of fuel consumed), EFji, for each pollutant
j and for each fuel type i were estimated for each of the
four seasons of the year. The emission factors for CO,
CH4, NHMC, PM2.5, and PM10 were calculated using the
proposed formulas by Ward et al.,36 which are the ones
incorporated to the FOFEM smoke production module.
Combustion efficiency, CEji, for each pollutant j and for
each fuel type i is a parameter that is incorporated in
the proposed formulas. The efficiencies used are an
average of the combustion efficiencies for the flaming
and smoldering phases of a fire, weighted by the fraction of each phase in the fire.23 Note that the definition
of combustion efficiency used by FOFEM is related exclusively to the fate of the carbon contained in the

burned mass, that is, mass of carbon released in the

form of CO2 divided by the total mass of carbon released. The emission factors for NOx were calculated
using a formulation that assumes emissions proportional to the nitrogen content of the fuel,37 with an
average fuel nitrogen of 0.7%.22 For NH3 emissions, a
molar ratio of NH3 to NOx corrected by a modified
combustion efficiency was used.38 Table 2 reports the
emission factors obtained from the approach described
for each fuel component. Additional details on these
calculations can be found elsewhere.22,23
Next, the consumed loads estimated with FOFEM 5
were combined with the emission factors shown in Table
2 to get a set of composite emission factors ([CEFs] lb of
pollutant/acre burned) as follows:
CEF jk

EF ji CL ik

(1)
where CEFjk represents the CEF
of the jth pollutant species and
the kth vegetation type. As an example, Table 3 reports the CEFs
obtained for Whitebark Pine
cover type during springtime.
Finally, the total emissions
for each pollutant (Ej) were calculated using the CEFs and the
total area burned for each vegetation type (Ak):
Ej
Figure 4. Number of wildfires per month by state.
1802 Journal of the Air & Waste Management Association

CEF jk A k

(2)

Volume 55 December 2005

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Mendoza, Garcia, Vela, Lozano, and Allen

RESULTS AND
DISCUSSION
A summary of the emissions inventory is shown in Table 4. It
includes wildfires and agricultural burning estimated for
northeastern Mexico for the period of January to August of
2000. The results indicate that
50% of the fires occurred in
the state of Tamaulipas, representing about the same percentage contribution in burned
area. The state of Coahuila was
affected by 30% of the fires,
leaving the state of Nuevo Leon
with the lowest contribution in
burned area. With respect to
the total emissions by state, Figure 5. Total burned area per month by state.
Coahuila contributed with
58.5%, Tamaulipas with 30.5%, and Nuevo Leon with
only 31% of the fires in Tamaulipas affected vegetation
11%. The results indicate that although Tamaulipas was
with high CEFs.
Figures 4 and 5 present the total number of fires and
the state with the highest number of fire occurrences and
area
burned, respectively, for each month, and by state, in
area burned, it was not the one with the most emissions.
the
period
of interest. Most fires are reported between
This is because of the mix of vegetation types burned in
April
and
May,
with a very drastic reduction in fires dureach state. Figure 3 reports the estimated burned area by
ing
the
rain
season
(summer). This monthly distribution
the fires identified for the whole region for different vegis
consistent
with
the
historical behavior of fires in Mexetation types. Additional analysis of the results indicates
ico.21 As expected from the temporal distribution of fires,
that in Coahuila, of the total burned area, 37% was mixed
most of the emissions were generated during the months
desert scrub, 20% Chaparral, 15% desert scrub, and 7%
of April and May, as seen in Figure 6. Emissions for each
mosaic cropland. Nuevo Leon was preferentially affected
pollutant by month and by state are not presented for
in lands covered by Tamaulipan thornscrub (18%), shrub
brevity, although they follow a very similar temporal
steppe (17%), mosaic cropland (16%), cultivated grasspattern as the one shown in Figure 6, with small differlands (12%), mixed desert scrub (12%), and mesquite
ences from state to state. Also, the monthly contribution
shrub (8%). Finally, in Tamaulipas, most of the burned areas
were covered by deciduous seasonal forest (26%), mosaic
cropland (24%), cultivated
grassland (24%), shrub steppe
(8%), and Tamaulipan thornscrub (7%). Table 5 lists the
CEFs for the vegetation types
that were involved preferentially in the wildfire events. The
highest values are reported for
Tamaulipan thrornscrub and
deciduous seasonal forest, followed by mixed desert scrub,
desert scrub, and shrub steppe.
In Coahuila, 52% of the fires
occurred in either mixed desert
scrub or desert scrub, whereas Figure 6. Total emissions per month by state.
Volume 55 December 2005

Journal of the Air & Waste Management Association 1803

Mendoza, Garcia, Vela, Lozano, and Allen

Downloaded by [201.170.123.1] at 13:49 21 November 2014

of each pollutant to the total emissions by state varies

little with respect to the average contribution that can
be derived from Table 1. Details on this information
can be found elsewhere.39 Finally, details on the spatial
and temporal distribution of wildfires are illustrated in
Figure 7.
The total load of wildfire emissions to the atmosphere
was compared with the official 1995 emissions inventory
of the city of Monterrey, in the state of Nuevo Leon, as a
reference of the magnitude of the contribution of wildfires to the total load of pollutants in the atmosphere of
the region. Monterrey is the most important industrial,
commercial, and cultural center in Northeastern Mexico.
It has a population exceeding three million inhabitants.

The official emissions inventory for Monterrey40 reports annual emissions of 898,268 short tons of total
suspended particles, 999,735 short tons of CO, 58,672
short t of NOx, and 138,078 short t of total hydrocarbons (THCs). No estimates of NH3 are available, and the
THCs do not include biogenic contributions. The comparison of these figures to the emissions from wildfires
yields ratios that go from 1.91:1 (Monterrey emissions
versus total emissions from wildfires) for CO to as high
as 13.7:1 for NOx, indicating that the region-wide emissions from wildfires are lower (and, in some cases, considerably lower) than the ones observed in the referenced urban center. An additional compassion was made
between the emissions estimated here and the emissions
estimated for the years 1996
and 199722 for the state of
Texas (Table 6), obtaining similar figures for both regions.

Figure 7. Classification of fire pixels by month between January and August 2000.
1804 Journal of the Air & Waste Management Association

Wind Trajectories
Analysis
An emissions inventory for
major pollutant species because of wildfires was generated for three northeastern
Mexican states. The emissions
inventory by itself is a valuable tool to account for the
relative contribution of these
emissions to the air quality in
the region. Here, we additionally investigated the possible
transboundary transport of
those emissions from Mexico
to the United Sates.
For 2000, prevailing wind
patterns at 850 mb obtained
from National Centers for Environmental Prediction reanalysis data41 in the region
shifted from westerly and
southerly winds late in the
winter and early in the spring
to southerly and southeasterly late in the spring and
early in the summer (Figure
8). The 850-mb level was selected, because it tends to accurately represent long-range
transport of pollutants in the
troposphere. Transport was
characterized by slow-moving
air masses during the first
Volume 55 December 2005

Mendoza, Garcia, Vela, Lozano, and Allen

Table 6. Comparison of emissions estimated in northeastern Mexico in
this study to annual emission estimates for the state of Texas for the
years 1996 and 1997.
Emissions (short t)

Pollutant

Downloaded by [201.170.123.1] at 13:49 21 November 2014

CO
CH4
NMHC
PM10
PM2.5
NOx
NH3

Northeastern
Mexico

Texasa

506,000
24,000
32,200
41,900
49,400
4290
5240

520,000
24,000
36,500
52,500
45,000
7450
5650

Notes: CH4 methane; NMHC nonmethane hydrocarbon; aAverage of the

annual emissions estimated for 1996 and 1997.

quarter of the year, a situation that changed in the spring

when winds tended to strengthen. These patterns have
been documented by other authors for other years as
well.18,20
Wind forward-trajectories simulations were done to
complement this preliminary transport analysis. The
Hybrid-Single Particle Lagrangian Integrated Trajectory
(HYSPLIT) model42 was used for this purpose. HYSPLIT
was run on the Realtime Environmental Applications
and Display System website.43 HYSPLIT uses a hybrid
method between a Lagrangian approach for transport
calculations and a Eulerian approach for dispersion calculations.
HYSPLIT was used to study wind trajectories for the
largest fires in the inventory. Archive data from reanalysis from 1996 to August 2001 stored in the HYSPLIT
website were used in the computations. Three starting

Figure 8. Mean wind direction and speed (m/sec) at 850 mb for 2000: (a) JanuaryFebruary, (b) MarchApril, (c) MayJune, and (d)
JulyAugust. Images obtained through the National Oceanic and Atmospheric Administration-CIRES Climate Diagnostics Center, Boulder, CO
(http://www.cdc.noaa.gov/).
Volume 55 December 2005

Journal of the Air & Waste Management Association 1805

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Mendoza, Garcia, Vela, Lozano, and Allen

Figure 9. Forward trajectories for wildfires occurring: (a) March 7, (b) March 29, (c) March 30, (d) April 23, (e) May 11, (f) May 11, and (g) June
1, at different locations in Northeastern Mexico.

heights were used: 10,500 and 1000 m above ground level.

Each run was performed for 72 hr forward trajectory.
HYSPLIT was run under the model vertical velocities mode.
Figure 9 reports eight selected forward trajectories for
air masses passing along the sites where major fires occurred between March and June of 2000. In all of the
cases, the 500-m and 1000-m forward trajectories end
transporting the air masses to and across the international
border. These heights represent expected plume rise distances; thus, there is a high probability that these trajectories transported air pollutants generated from wildfires
in northeastern Mexico to the United States.
1806 Journal of the Air & Waste Management Association

CONCLUSIONS
Previous work has shown that emissions from wildfires
and prescribed agricultural burning can have an important contribution in the degradation of air quality. A first
step to obtain the contribution and impacts of such
events is to generate an emissions inventory from that
source. Here, an inventory of air pollutants emitted from
wildfires in northeastern Mexico between January and
August of 2000 was created. The main source of information used to create the inventory came from satellite imagery. Surveys conducted indicated that prescribed agricultural burning is not an important contributor to the
Volume 55 December 2005

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Mendoza, Garcia, Vela, Lozano, and Allen

Figure 9. Cont.

total emissions, although cultivated grasslands and mosaic croplands were identified through satellite imagery as
being routinely affected by fires (presumably wildfires).
Overall, the emissions estimated from wildfires for northeastern Mexico are comparable with figures obtained for
the state of Texas, in the United States. Coahuila was the
state that contributed with most of the total estimated
emissions for the region (58%), although it only contributed 31% of the total area burned. This discrepancy is
because of the mix of vegetation types burned in each
state: preferentially, shrub types with high-emission factors in Coahuila and deciduous forest, cropland, and
grasslands with lower-emission factors in Tamaulipas.
Temporal distribution was in line with expectations: April
and May were the months where more fires occurred.
Volume 55 December 2005

Preliminary analysis of wind forward trajectories of air

masses passing through the areas burned indicate possible
transboundary transport of the emissions from Mexico to
the United States during the occurrence of the major wildfires identified. This is verified by the observation of average
wind patterns obtained from wind reanalyses data.
ACKNOWLEDGMENTS
Support for this project came from Agreement UTA03
261 between the University of Texas and the Instituto
Tecnologico y de Estudios Superiores de Monterrey.
REFERENCES
1. Auclair, A.N.D., Carter, T.B. Forest Wildfires as a Recent Source of
Carbon Dioxide at Northern Latitudes; Can. J. Forest Res. 1993, 23,
1528 1536.
Journal of the Air & Waste Management Association 1807

Downloaded by [201.170.123.1] at 13:49 21 November 2014

Mendoza, Garcia, Vela, Lozano, and Allen

2. Conard, S.G.; Ivanova, G.A. Wildfire in Russian Boreal Forests-Potential Impacts of Fire Regime Characteristics on Emissions and Global
Carbon Balance Estimates; Environ. Pollut. 1998, 98, 305313.
3. French, N.H.F.; Kasischke, E.S.; Williams, D.G. Variability in the Emissions of Carbon-Based Trace Gases from Wildfire in the Alaskan Boreal
Forest; J. Geophys. Res. 2003, 108, 8151 8161.
4. Mazurek, M.A.; Cass, G.R.; Simoneit, B.R.T. Biological Input to Visibility-Reducing Particles in the Remote Arid Southwestern United States;
Environ. Sci. Technol. 1991, 25, 684 694.
5. Watson, J.G. Visibility: Science and Regulation; J. Air & Waste Manage.
Assoc. 2002, 52, 628 713.
6. Park, R.J.; Jacob, D.J.; Chin, M.; Martin, R.V. Sources of Carbonaceous
Aerosols Over the United States and Implications for Natural Visibility;
J. Geophys. Res. 2003, 108, 4355 4368.
7. Chung, Y.S. On the Forest Fires and the Analysis of Air Quality Data
and Total Atmospheric Ozone; Atmos. Environ. 1984, 18, 21532157.
8. Wofsy, S.C.; Sachse, G.W.; Gregory, G.L.; Blake, D.R., Bradshaw, J.D.;
Sandholm, S.T.; Singh, H.B.; Barrick, J.A.; Harriss, R.C.; Talbot, R.W.;
Shipman, M.A.; Browell, E.V.; Jacob, D.J.; Logan, J.A. Atmospheric
Chemistry in the Artic and Subartic: Influence of Natural Fires, Industrial Emissions, and Stratospheric Inputs; J. Geophys. Res. 1992, 97,
1673116746.
9. McKeen, S.A.; Wotawa, G.; Parrish, D.D.; Holloway, J.S.; Buhr, M.P.;
Hubler, G.; Fehsenfeld, F.C.; Meagher, J.F. Ozone Production from
Canadian Wildfires during June and July of 1995; J. Geophys. Res.
2002, 107, 4192 4216.
10. Duncan, B.N.; Bey, I.; Chin, M.; Mickley, L.J.; Fairlie, T.D.; Martin,
R.V.; Matsueda, H. Indonesian Wildfires of 1997: Impact on Tropospheric Chemistry; J. Geophys. Res. 2003, 108, 4458 4483.
11. Friedli, H.R.; Radke, L.F.; Lu, J.Y.; Banic, C.M.; Leaitch, W.R.; MacPherson, J.I. Mercury Emissions from Burning of Biomass From Temperate
North American Forests: Laboratory and Airborne Measurements; Atmos. Environ. 2003, 37, 253267.
12. Manoe, S.; Andreae, M.O. Emission of Methyl Bromide from Biomass
Burning; Science 1994, 263, 12551257.
13. Hsu, N.; Herman, J.; Bhartia, P.; Seftor, C.; Torres, O.; Thompson, A.;
Gleason, J.; Eck, T.; Holben, B. Detection of Biomass Burning Smoke
from TOMS Measurements; Geophys. Research Lett. 1996, 23, 745748.
14. Goode, J.G.; Yokelson, R.J.; Ward, D.E.; Susott, R.A.; Babbitt, R.E.;
Davies, M.A.; Hao, W.M. Measurements of Excess O3, CO2, CO, CH4,
C2H4, C2H2, HCN, NO, NH3, HCOOH, CH3COOH, HCHO, and
CH3OH in 1997 Alaskan Biomass Burning Plumes by Airborne Fourier
Transform Infrared Spectroscopy (AFTIR); J. Geophys. Res. 2000, 105,
2214722166.
15. Fraser, M.P.; Lakshmanan, K. Using Levoglucosan as a Molecular
Marker for the Long-Range Transport of Biomass Combustion Aerosols; Environ. Sci. Technol. 2000, 34, 4560 4564.
16. Conny, J.M.; Slater, J.F. Black Carbon and Organic Carbon in Aerosol
Particles from Crown Fires in the Canadian Boreal Forest; J. Geophys.
Res. 2002, 107, 4116 4127.
17. Hays, M.D.; Geron, C.D.; Linna, K.J.; Smith, N.D.; Schauer, J.J. Speciation of Gas-Phase and Fine Particle Emissions from Burning of Foliar
Fuels; Environ. Sci. Technol. 2002, 36, 22812295.
18. Gebhart, K.A.; Kreidenweis, S.M.; Malm, W.C. Back-Trajectory Analyses of Fine Particulate Matter Measured at Big Bend National Park in
the Historical Database and the 1996 Scoping Study; Sci. Total Environ.
2001, 276, 185204.
19. Kreidenweis, S.M.; Remer, L.A.; Bruintjes, R.; Dubovik, O. Smoke Aerosol from Biomass Burning in Mexico: Hygroscopic Smoke Optical
Model; J. Geophys. Res. 2001, 106, 4831 4844.
20. Peppler, R.A.; Bahrmann, C.P.; Barnard, J.C.; Campbell, J.R.; Cheng,
M.-D.; Ferrare, R.A.; Halthore, R.N.; Heilman, L.A.; Hlavka, D.L.; Laulainen, N.S.; Lin, C.-J.; Ogren, J.A.; Poellot, M.R.; Remer, L.A.; Sassen,
K.; Spinhirne, J.D.; Splitt, M.E.; Turner, D.D. Arm Southern Great
Plains Site Observations of the Smoke Pall Associated with the 1998
Central American Fires; Bull. Am. Meteorol. Soc. 2000, 81, 25632591.
21. Bravo, H.A.; Sosa, E.R.; Sanchez, A.P.; Jaimes, P.M.; Saavedra, R.M.I.
Impact of Wildfires on the Air Quality of Mexico City, 19921999;
Environ. Pollut. 2002, 117, 243253.
22. Dennis, A.; Fraser, M.; Anderson, S.; Allen, D. Air Pollutant Emissions
Associated with Forest, Grassland, and Agricultural Burning in Texas;
Atmos. Environ. 2002, 36, 3779 3792.
23. Dennis, A. Inventory of Air Pollutant Emissions Associated with Forest,
Grassland and Agricultural Burning in Texas; M.S. Thesis, University of
Texas at Austin, Austin, TX, May 2000.
24. Allen, D.; Katamreddy, A.; Junquera, V.; Decabooter, J. Emissions Inventory Associated with Forest, Grassland, and Agricultural Burning during
the Texas Air Quality Study; Texas Environmental Research Consortium
Project H-3 Final Report; University of Texas: Austin, TX, 2002.
25. Saunders, R.W.; Kriebel, K.T. An Improved Method for Detecting Clear
Sky and Cloudy Radiances from AVHRR Data; Int. J. Rem. Sensing 1988,
9, 123150.
26. Uso de Suelo y Vegetacion Modificado por CONABIO; Escala 1: 1000 000;
Comision Nacional para el Conocimiento y Uso de la Biodiversidad
(CONABIO): Ciudad de Mexico, Mexico, 1999.

1808 Journal of the Air & Waste Management Association

27. Molina, E. Department of Agriculture and Food Technology, Instituto

Tecnologico y de Estudios Superiores de Monterrey, Monterrey, Mexico. Personal communication, 2003.
28. Reinhardt, E. Using FOFEM 5.0 to Estimate Tree Mortality, Fuel Consumption, Smoke Production, and Soil Heating from Wildland Fire; USDA Forest
Service, Missoula Fire Sciences Laboratory: Missoula, MT, 2003.
29. Albini, F.A.; Reinhardt, E.D. Modeling Ignition and Burning Rate of
Large Woody Natural Fuels; Int. J. Wild. Fire 1995, 5, 8191.
30. Albini, F.A.; Reinhardt, E.D. Improved Calibration of a Large Fuel
Burnout Model; Int. J. Wild. Fire 1997, 7, 2128.
31. Barbour, M.G.; Billings, W.D. North American Terrestrial Vegetation.
Cambridge University Press: Cambridge, UK, 1988.
rboles y Arbustos de la
32. Castillo-Campos, G.; Medina-Abreo, M.E. A
Reserva Natural de la Mancha, Veracruz; Manual para la Identificacion
de las Especies; Instituto de Ecologa A.C.: Xalapa, Veracruz, 2002.
33. Rojas-Mendoza, P. Generalidades sobre la Vegetacion del Estado de Nuevo
Leon y Datos Acerca de su Flora; Doctoral Thesis, Universidad Nacional
Autonoma de Mexico, Facultad de Ciencias, Departamento de Biologa, Mexico, DF, 1965.
34. International Classification of Ecological Communities: Terrestrial Vegetation of Southeastern United States; The Natural Conservancy, Southern
Conservation Science Department, Community Ecology Group:
Chapel Hill, NC, 1998.
35. Mendoza, A.; Garcia, M.; Vela, P.; Lozano, F.; Allen, D. Emissions
Inventory from Wildfires and Prescribed Agricultural Activities in
Northeastern Mexico during Spring and Summer of 2000. In Proceedings of the 97th Annual Conference of the Air & Waste Management
Association; Air & Waste Management Association: Indianapolis, IN,
2004: Paper 436.
36. Ward, D.E.; Peterson, J.; Hao, W.M. An Inventory of Particulate Matter
and Air Toxic Emissions from Prescribed Fires in the USA for 1989; Report
A1299; USDA Forest Service, Intermountain Research Station: Missoula, MT, 1993.
37. Dignon, J.; Penner, J.E. Biomass Burning: A Source of Nitrogen Oxides
in the Atmosphere. In Global Biomass Burning; Vol. 1, Chapter 4; MIT
Press: Cambridge, MA, 1991, pp 370 375.
38. Yokelson, R.J.; Susott, R.A.; Babbitt, R.E.; Hao, W.M.; Ward, R.E. Trace
Gas Emissions from Specific Biomass Fire-Types. In Proceedings of the
START Synthesis Workshop on Greenhouse Gas Emission, Aerosol and
Land Use and Cover Change in Southeast Asia; Monya T.B., Ed. Southeast
Asian Regional Committee for START: Bangkok, Thailand, 1997.
39. Mendoza, A.; Lozano, D.F.; Garcia, M.R.; Vela, P.; Leal, S. Emissions
Inventory Associated with Wildfires and Prescribed Agricultural Burning in
Northeastern Mexico during Spring and Summer of 2000; Final Report to
the Center for Energy and Environmental Resources, The University of
Texas at Austin, Agreement UTA03261; Instituto Tecnologico y de
Estudios Superiores de Monterrey: Monterrey, Mexico, 2004.
40. Gestion de la Calidad del Aire en Mexico: Logros y Retos para el Desarrollo
Sustentable 19952000; Instituto Nacional de Ecologa, Direccion General de Gestion e Informacion Ambiental: Mexico, DF, 2000.
41. Kalnay, E.; Kanamitsu, M.; Kistler, R.; Collins, W.; Deaven, D.; Gandia,
L.; Iredell, M.; Saha, S.; White, G.; Woollen, J. et al.: The NCEP/NCAR
Reanalysis 40-Year Project; Bull. Amer. Meteor. Soc. 1996, 77, 437 471.
42. Draxler, R.R. HYSPLIT_4 Users Guide: Version 4.6 Last Revision October 2003; NOAA Technical Memorandum ERL ARL-230; National Oceanic & Atmospheric Administration: Silver Spring, MD, 1999.
43. Realtime Environmental Applications and Display System, available at
http://www.arl.noaa.gov/ready.html (accessed July 5, 2005).

About the Authors

Alberto Mendoza is an Associate Professor in the Department
of Chemical Engineering, Instituto Tecnologico y de Estudios
Superiores de Monterrey, Monterrey, Nuevo Leon, Mexico.
Marisa Garcia is a graduate student, Patricia Vela a research
associate, and Fabian Lozano a professor, all in the Center for
Environmental Quality, Instituto Tecnologico y de Estudios
Superiores de Monterrey. David Allen is the Henry Beckman
Professor in Chemical Engineering in the University of Texas
at Austin. Address correspondence to: Alberto Mendoza, Department of Chemical Engineering, Instituto Tecnologico y
de Estudios Superiores de Monterrey, Monterrey, Mexico
64849; phone: 52 81-8328 4336; fax: 52 81-8328
4144; e-mail: mendoza.alberto@itesm.mx.

Volume 55 December 2005