BIOMASS BURNING: PARTICLE EMISSIONS, … · BIOMASS BURNING: PARTICLE EMISSIONS, CHARACTERISTICS,...
Transcript of BIOMASS BURNING: PARTICLE EMISSIONS, … · BIOMASS BURNING: PARTICLE EMISSIONS, CHARACTERISTICS,...
QUEENSLAND UNIVERSITY OF TECHNOLOGY SCHOOL OF PHYSICAL AND CHEMICAL SCIENCES
BIOMASS BURNING: PARTICLE EMISSIONS,
CHARACTERISTICS, AND AIRBORNE
MEASUREMENTS Submitted by Arinto Yudi Ponco WARDOYO to the School of Physical and Chemical Sciences, Queensland University of Technology, in partial fulfilment of the requirements of the degree of Doctor of Philosophy.
July 2007
KEYWORDS
Biomass burning, emission factors, ultrafine particles, particle number emission,
particle size distribution, particle vertical profile, Queensland trees, Northern
Territory of Australia, particle number concentration, Northern Territory Australia,
airborne measurements, vertical profile.
ii
ABSTRACT
Biomass burning started to attract attention since the last decade because of its
impacts on the atmosphere and the environmental air quality, as well as significant
potential effects on human health and global climate change. Knowledge of particle
emission characteristics from biomass burning is crucially important for the
quantitative assessment of the potential impacts. This thesis presents the results of
study aimed towards comprehensive characterization of particle emissions from
biomass burning. The study was conducted both under controlled laboratory
conditions, to quantify the particle size distribution and emission factors by taking
into account various factors which may affect the particle characteristics, and in the
field, to investigate biomass burning processes in the real life situations and to
examine vertical profile of particles in the atmosphere.
To simulate different environmental conditions, a new technique has been developed
for investigating particle emissions from biomass burning in the laboratory. As
biomass burning may occur in a field at various wind speeds and burning rates, the
technique was designed to allow adjustment of the flow rates of the air introduced
into the chamber, in order to control burning under different conditions. In addition,
the technique design has enabled alteration of the high particle concentrations,
allowing conducting measurements with the instrumentations that had the upper
concentration limits exciding the concentrations characteristic to the biomass
burning.
iii
The technique was applied to characterize particle emissions from burning of several
tree species common to Australian forests. The aerosol particles were characterized
in terms of size distribution and emission factors, such as PM2.5 particle mass
emission factor and particle number emission factor, under various burning
conditions. The characteristics of particles over a range of burning phases (e.g.,
ignition, flaming, and smoldering) were also investigated. The results showed that
particle characteristics depend on the type of tree, part of tree, and the burning rate.
In particular, fast burning of the wood samples produced particles with the CMD of
60 nm during the ignition phase and 30 nm for the rest of the burning process. Slow
burning of the wood samples produced large particles with the CMD of 120 nm, 60
nm and 40 nm for the ignition, flaming and smoldering phases, respectively. The
CMD of particles emitted by burning the leaves and branches was found to be 50 nm
for the flaming phase and 30 nm for the smoldering phase, under fast burning
conditions. Under slow burning conditions, the CMD of particles was found to be
between 100 to 200 nm for the ignition and flaming phase, and 50 nm for the
smoldering phase.
For fast burning, the average particle number emission factors were between 3.3 to
5.7 x 1015 particles/kg for wood and 0.5 to 6.9 x 1015 particles/kg for leaves and
branches. The PM2.5 emission factors were between 140 to 210 mg/kg for wood and
450 to 4700 mg/kg for leaves and branches. For slow burning conditions, the average
particle number emission factors were between 2.8 to 44.8 x 1013 particles/kg for
wood and 0.5 to 9.3 x 1013 particles/kg for leaves and branches, and the PM2.5
iv
emissions factors were between 120 to 480 mg/kg for wood and 3300 to 4900 mg/kg
for leaves and branches.
The field measurements were conducted to investigate particle emissions from
biomass burning in the Northern Territory of Australia over dry seasons. The results
of field studies revealed that diameters of particles in ambient air emissions were
within the size range observed during laboratory investigations. The laboratory
measurements found that the particles released during the controlled burning were of
a diameter between 30 and 210 nm, depending on the burning conditions. Under fast
burning conditions, smaller particles were produced with a diameter in the range of
30 to 60 nm, whilst larger particles, with a diameter between 60 nm and 210 nm,
were produced during slow burning. The airborne field measurements of biomass
particles found that most of the particles measured under the boundary layer had a
CMD of (83 ± 13) nm during the early dry season (EDS), and (127 ± 6) nm during
the late dry season (LDS). The characteristics of ambient particles were found to be
significantly different at the EDS and the LDS due to several factors including
moisture content of vegetation, location of fires related to the flight paths, intensity
of fires, and burned areas. Specifically, the investigations of the vertical profiles of
particles in the atmosphere have revealed significant differences in the particle
properties during early dry season and late dry season. The characteristics of particle
size distribution played a significant role in these differences.
v
ACKNOWLEDGMENTS
In the Name of Allah, the most Gracious, the most Merciful, Praise to be Allah, Lord
of the Universe, and peace and prayer to upon His Final Prophet and Messenger.
Because of Allah, I have finished doing researches and writing a thesis for my study.
I wish to express my sincere gratitude, respect, honor, and appreciation to my
principal supervisor Professor Lidia Morawska for her kindness, help, patience,
guidance, enthusiastic support, expert feedback, and giving me a chance to join her
laboratory. Lidia has been not only my supervisor and mentor, but also like mother to
me, and sister, and a good friend over the years. I have learnt many things from Lidia
for my professional career.
I would also like to take this opportunity to thank Dr. Zoran Ristovoski, my co-
supervisor, for his guidance and expert feedback. Zoran has made important
contributions to this work, especially for the airborne measurements.
I would like to thank Dr. Jack Marsh and Dr. Riaz Akbar for their assistance during
the sample collection.
I would like to thank Dr. Milan Jamriska for his contribution to the airborne
measurements and Dr. Graham Johnson for preparing the equipment for the airborne
measurements and his assistance with instrumentation problem solving.
I would also like to thank Dr. Congrong He for his assistance with calibrating
equipment and Dr. Victoria Agronovski for her assistance with thesis editing and
advices.
vi
Many thanks to Chris Duncan for his assistance in purchasing a stove and
suggestions in designing a burning system, and Stuart Costin for his assistance with a
computer support.
I thank also Rachael Robson for administrative assistance and correcting my papers
and Gillian Isoardi for language correction of my paper and thesis.
Special thanks go to Husien, Sade, Afkar, and Galina for being good friends, for their
humor, joking, and criticism. I have had a wonderful time.
To my lovely wife Kartika, thank so much for your pain, faith, sacrificing,
understanding, support, and patience over these years. To my wonderful daughter
Eva, many thanks for your understanding and being good.
vii
LIST OF PUBLICATIONS
Wardoyo, A.Y.P., Morawska, L, Ristovski, Z., and Marsh, J. (2006).
Quantification of particle number and mass emission factors from combustion
of Queensland trees. Journal Environmental Science and Technology, 40,
5696-5703.
Ristovski, Z., Wardoyo, A.Y.P., Morawska, L, Jamriska, M., Carr, S., and Johnson,
G. (2007). Biomass burning influenced particle characteristics in Northern Territory
Australia based on airborne measurements. Submitted for publication in Journal of
Geophysical Research.
Wardoyo, A.Y.P., Morawska, L, Ristovski, Z., Jamriska, M., Carr, S., and Johnson,
G. (2007). Size distribution of particle emitted from grass fires in the Northern
Territory Australia. Submitted for publication in Atmospheric Environment.
viii
TABLE OF CONTENTS
Keywords …………………………………………………...…….…… ii
Abstract ……………………………………………………………….… iii
Acknowledgements ………………………………………………..… vi
List of publications ……………………………………………………... viii
List of tables ……………………………………………………….….. xvi
List of figures …………………………………………………………… xvii
Statement of original authorship ………………………………………... xix
CHAPTER 1. INTRODUCTION …………………………………….. 1
1.1. Description of research problem investigated ……………. 1
1.2. Motivation of the study …………………………………… 4
1.3. Overall objective of the study …………………………..... 5
1.4. Specific aims of the study …………………………………. 5
1.5. Account of research progress linking the research papers … 6
1.6. References ………………………………………………. 10
CHAPTER 2. LITERATUTE REVIEW ……………………………..... 15
2.1. Introduction …………………………………………………… 15
2.2. Air quality. ……………………………………………………. 15
2.2.1. Airborne particles: general background and definitions 16
2.2.2. Particle size distribution ……………………………. 17
2.2.3. Emission factors ……………………………………. 19
2.2.4. Summary …………………………………………… 20
2.2.5. Concentrations of particulate matter in different countries 20
ix
2.2.6. Summary . ……………………….…………………. 22
2.2.7. Ambient air quality standard ………………………. 22
2.2.8. Summary …………………………………………… 24
2.2.9. Gaseous compounds in different countries ………... 25
2.3. Biomass burning: general background ……………….……… 26
2.3.1. Definitions ………………………………………… 27
2.3.2. Summary …………………………………………… 29
2.3.3. Area of biomass burning …………………….…. 30
2.3.4. Summary …………………………………………… 30
2.3.5. Biomass burning emissions …………………………. 31
2.3.5.1. Emission rate….. ………………………….. 31
2.3.5.2. Type of emissions ………………………. 31
2.3.6. Summary ……………………………………………. 36
2.4. Particles originating from biomass burning …………………… 36
2.4.1. Particle formation ……………………………………. 36
2.4.2. Particle composition ……… ………………………… 37
2.4.3. Summary …………………………………………….. 38
2.4.4. Characteristic of particle size ……….……………… 38
2.4.5. Summary ……………………………………………. 41
2.4.6. Particle emission factors …………………………….. 41
2.4.7. Summary ……………………………………………. 43
2.5. Measurements of biomass burning particles ………………..… 44
2.5.1. Fresh particles ………………………………….…… 44
2.5.2. Aged particles …………………………….………… 45
x
2.5.3. Particle measurement methods …………………….. 45
2.5.4. Summary ……………………………………………. 47
2.6. Biomass Burning in Australia ………………………….…….. 48
2.6.1. Summary ……………………………………………. 49
2.7. Biomass burning health impacts ………….………………….. 50
2.7.1. Particulate matter ……………………….………….. 51
2.7.2. Summary ……………………………………………. 55
2.7.3. Polycyclic aromatic hydrocarbons (PAHs) ………….. 55
2.7.4. Carbon monoxide ……………………………………. 56
3.7.5. Aldehydes ……………………………………………. 57
2.7.6. Organic acids ………………………………………… 57
2.7.7. Volatile organic compounds …………………………. 58
2.7.8. Dioxin ………………………………………………... 58
2.7.9. Elementary compounds ………………………………. 59
2.7.10. Summary ……………………………………………. 59
2.8. Dispersion model……………………………………………….. 59
2.8.1. Theoretical background ………………………………. 59
2.8.2. Classification dispersion model ………………………. 61
Lagrangian model ……...…………………………… 62
Eularian model ……………………………………… 68
Statistical particle model ……..……………………. 71
2.8.3. Dispersion model for biomass burning ………………… 72
2.8.4. A model for biomass burning study …………………… 75
2.8.4.1. Strength source models ……………………… 75
xi
2.8.4.2. Meteorological models ……………...………. 77
2.8.4.3. Dispersion model approach …………………. 78
Objective of study ……….. …………………. 79
Physical processes ………..………………… 79
Complete model ………..…………………… 80
Consistency of model ………………………… 82
High resolution ………..…………………….. 82
Brisbane terrain ………..……………………. 83
2.8.5. Summary ………………………………………. 83
2.9. Conclusions ……………………………………………………… 84
2.10. Knowledge Gaps in regards to particle characteristics from biomass
burning in general and in Australia especially……………..……………. 87
2.11. References …………………….…………….……………………. 91
CHAPTER 3. QUANTIFICATION OF PARTICLE NUMBER AND MASS
EMISSION FACTORS FROM COMBUSTION OF QUEENSLAND TREES
3.1. Introduction …………………………………………….………... 132
3.2. Experiment section …………………………………………….... 133
3.2.1. Experiment setup …………………………………….... 134
3.2.2. Burning system ……………………………………….. 135
3.2.3. Particle measurement system …………………………. 136
3.2.4. Dilution and sampling system ………………………… 137
3.2.5. Sample material and preparation ……………………… 138
3.2.6. Burning conditions ……………………………………. 140
3.3. Result and discussion …………………………………………… 142
xii
3.3.1. Sampling system performance ………………………… 142
3.3.2. Particle size distribution ……………………………….. 142
3.3.3. Particle number concentrations ………………………... 145
3.3.4. PM2.5 concentrations …………………………………... 148
3.3.5. Particle number emission factors ……………………… 148
3.3.6. PM2.5 emission factors ………………………………… 151
3.4. References ………………………………………………………. 153
CHAPTER 4. BIOMASS BURNING INFLUENCED PARTICLE
CHARACTERISTICS IN NORTHERN TERRITORY AUSTRALIA FROM
AIRBORNE MEASUREMENTS
4.1. Introduction ……………………………………………………… 164
4.2. Experiment methods ……………………………………………. 166
4.2.1. Study area …………………………………………….. 166
4.2.2. Measurement times and locations …………………….. 167
4.2.3. Instrumentation setup …………………………………. 170
4.3. Result and discussion …………………………………………… 172
4.3.1. Temperature and relative humidity …………………… 172
4.3.2. Height of the boundary layer …………………………. 173
4.3.3. Particle concentrations during June campaign ………… 174
4.3.4. Particle concentrations during September campaign ….. 180
4.4. Discussion and conclusion ……………………………………… 183
4.5. References ………………………………………………………. 187
xiii
CHPTER 5. LABORATORY AND AIRBORNE MEASUREMENTS OF SIZE
DISTRIBUTION OF PARTICLE EMISSION FROM BIOMASS BURNING
5.1. Introduction …………………………………………………….. 196
5.2. Methods ………………………………………………………… 198
5.2.1. Laboratory measurements ……………………………. 198
5.2.1.1. Experiment setup …………………………… 199
5.2.1.2. Sample material and preparation …………… 200
5.2.1.3. Burning conditions …………………………. 201
5.2.2. Airborne measurements ………………………………. 201
5.2.2.1. Study area …………………………………… 202
5.2.2.2. Measurement time and location …………….. 203
5.2.3. Data analysis ………………………………………….. 204
5.3. Results …………………………………………………………… 205
5.3.1. Laboratory measurements …………………………….. 205
5.3.1.1. Particle size distributions ……………………. 205
5.3.2. Airborne measurements ………………………………… 207
5.3.2.1. Boundary layer measurements …………..…… 207
5.3.2.2. Particle size distributions ……..……………… 208
5.4. Discussion and conclusion ……………………………………… 210
5.4.1. Particle diameter ……………………………………… 210
5.4.1.1. Laboratory studies ………………………….. 211
5.4.1.2. Field studies ………………………………… 214
5.4.1.3. Comparison between CMD measured in the laboratory
and in the field ………………….…………. 215
xiv
5.4.2. Particle vertical profile ……………………………….. 216
5.4.3. Particle age ……………………………………………. 217
5.5. References ………………………………………………………. 220
CHAPTER 6. GENERAL DISCUSSION ………………………………… 227
6.1. Introduction ……………………………………………………… 227
6.1.1. Biomass burning emissions …………………………… 227
6.1.2. Biomass burning particles …………………………… 228
6.1.3. Biomass burning impacts ……………………………. 230
6.1.4. Characteristics of biomass burning particles ………… 233
6.2. Principal significance of findings ……………………………… 236
6.3. Conclusions …………………………………………………….. 244
6.4. Scientific recommendations …………….……………………. .. 244
xv
LIST OF TABLES
2.1 Ambient air quality guideline for particulate matter.
2.2 Characteristics of the dispersion models for biomass burning.
4.1 Summary of measurement flight plans.
5.1 Particle concentration measured during the campaigns.
xvi
LIST OF FIGURES
2.1 The concentration of PM2.5 in several cities around the world.
2.2 Concentration of PM2.5 and PM10 in several cities of Australia.
2.3 PM2.5 emission factor from wood burning.
2.4 The relationship between the PM10 concentration and morbidity and
mortality.
3.1 Experimental setup consisting of the burning system (modified stove), a
dilution and sampling system, and a particle measurement system.
3.2 Representative size distribution of particles from fast burning of woods.
3.3 Count Median Diameter characteristic of samples burned.
3.4 Total number concentration for fast burning of Blue Gum.
3.5 Average particle number emission factors of burned samples for fast burning
and slow burning.
3.6 PM2.5 emission factors of burned samples.
4.1 Location of the flight tracks (source: http://www.sentinel.csiro.com.au). The
two maps zoomed in over the flight path (indicated by the black line) at the
Northern end of the Northern Territory, show satellite fire spot data for 22-28
June 2003 (left) and 21-27 September 2003 (right).
4.2 Temperature and relative humidity as a function of height for June and
September campaign.
4.3 The vertical temperature profile measured on the 26th of June.
4.4 The measured concentrations of particles for June campaign.
xvii
4.5 The particle concentrations along the flight paths, measured on 24th June
2003 at 800 m and 1800 m. The large triangles show the location of fires. The
arrows show wind directions at the noted heights.
4.6 Measured particle concentrations for September campaign.
4.7 (a) Average particle concentrations and (b) Count Median Diameter (CMD)
measured during June and September campaigns.
5.1 Savannas in the Northern Territory Australia with a variety of vegetation.
The black line indicates the flight path flown at various altitudes during the
campaigns.
5.2 The average of size distribution for slow burning of grass samples.
5.3 Count median diameter (CMD) characteristic of samples burned.
5.4 The temperature vertical profile measured on the 26th of June 2003.
5.5 Average size distribution for the June and September campaigns.
5.6 The measured Count Median Diameter of particles during June and
September campaigns.
6.1 Diagram of the research activities.
xviii
THE STATEMENT OF ORIGINAL AUTHORSHIP
This work contained in this thesis has not been previously submitted for a degree or
diploma at any other education institution. To be best of my knowledge and belief,
the thesis contains no material previously published or written by another person
except where due reference is made.
Signed: ……………………..
Date: ………………………..
xix
CHAPTER 1. INTRODUCTION
1.1 A DESCRIPTION OF SCIENTIFIC PROBLEM INVESTIGATED Biomass burning has attracted the attention of people around the world due to its
contribution of particles and gases in the atmosphere (Dennis et al., 2002; Uherek,
2004), and the impact of these emissions on human health, which have been linked to
morbidity and mortality (Dockery et al., 1993; Schwartz, 1993; Etzel, 1999;
McConnel et al., 1999; Peters et al., 1999; Levy et al., 2000; Peters et al., 2000;
Samet et al., 2000a; Samet et al., 2000b; Tolbert et al., 2000; Yu et al., 2000).
Further to this, these emissions also play a significant role in affecting atmospheric
processes (Bodhaine, 1983; Shaw, 1987), such as radiation balance (Wurzler and
Simmel, 2005) or acidification of clouds, rain and fog (Nichol, 1997). The impact on
the radiation balance of the earth occurs both directly, by absorbing and scattering
incoming solar radiation; and indirectly, by acting as cloud condensation nuclei
(CCN) and also by altering the clouds microphysical processes on a mesoscale
(Kaufman et al., 1998; Martins et al., 1998; Wurzler and Simmel, 2005).
Huge areas around the world have been affected by biomass burning and data shows
that 500 to 1000 million hectares of open forest and savannas, 1 million hectares of
forest in northern latitudes, and 4 million hectares of tropical and sub tropical forest
are burnt every year (Uherek, 2004). Biomass burning in Texas destroyed 0.5 million
hectares in 1996 and 1997 (Dennis et al., 2002) and more than 1.3 million hectares of
forest were burnt in China in 1987. In the same year, forest fires in eastern Asia
consumed approximately 14 million hectares (Cahoon et al., 1992) and forest
1
burning destroyed more than 20 million hectares in Indonesia in 1994 and 1997
(Nichol, 1997).
Biomass burning in Australia burns 40-130 million hectares of land annually (WHO,
2000), and the fires of 1998-1999 and 1999-2000 consumed areas of 31 and 71
million hectares respectively (Gill and Moore, 2005). The Western Australian
Department of Land Information reported that biomass burning across Australia,
from 1997 to 2003, destroyed an area of approximately 26 to 80 million hectares.
The greatest extent of biomass burning is reported in the savannas of the Northern
Territory of Australia. The Australian Bureau of Statistics (ABS) reported that, from
July 2002 to February 2003, there were 5,999 forest fires burning an area of 21
million hectares across Australia, with the majority located in the Northern Territory
of Australia, in an area of 15 million hectares (ABS, 2004). Biomass burning in the
savannas of the Northern Territory of Australia during the period of 1997-1999 was
estimated to affect an area of 30 million hectares (Russell-Smith et al., 2003a) and
during 1997-2001, an average of 30 million hectares of savannas in the Northern
Territory were affected by fires, with the greatest area damaged in 1999 when 4
million hectares were burned (Russell-Smith et al., 2003b). The data showed that
from 1980 to 1995, more than 1 million hectares of the Kakadu National Park in the
Northern Territory of Australia were destroyed by fires (Gill et al., 2000). The state
of Queensland experiences biomass burning every year and it was recorded that
Queensland fires in 1991 consumed 37,000 hectares of forest (Hamwood, 1992).
From July 2002 until June 2003, there were 2,618 fires destroying 1 million hectares
of forest in this state (ABS, 2004).
2
Biomass burning significantly contributes to the particle burden on the atmosphere.
Biomass burning produces total suspended particles (TSP) of 104 Tg/year and
contributes 38 % of particulate matter released by all sources every year (Andreae,
1991). Other data shows that burning of cereal wastes in Spain release a total
particulate matter (TPM) of 80 – 130 Gg annually (Ortiz de Zarate et al., 2000),
while wood burning in Sweden results in emission of 8,600 – 65,000 tonnes of
particulate matter every year (Areskoug et al., 2000).
Significant impacts of biomass burning on human health and atmospheric processes
have been recognized. Knowledge of characteristics of biomass burning particles has
been identified as a very important issue in developing quantitative assessment of the
impacts. Characterizing the nature of particle size distribution and particle emission
produced by biomass burning is important to assess the impacts on human health
associated with the depth of particle penetration into the lung and global changes
related to micro-physical processes in the atmosphere. The majority of particles
resulting from biomass burning has been reported with the diameter less than 2.5 μm
in the previous studies (Hays et al., 2002; Hedberg et al., 2002; Ferge et al., 2005;
Wieser and Gaegauf, 2005). PM2.5 emission factors due to biomass burning have
been measured in the range of 0.2 to 12 g/kg (McDonald et al., 2000; Fine et al.,
2002; Hays et al., 2002). In terms of particle number, the reported emission factors
for burning unspecific wood range from 3 x 1015 to 40 x 1016 particles/kg (Wieser
and Gaegauf, 2005). However, the existing data on particle size distribution and
number emission factors are still very limited, with unavailable data for many tree
species, such as those growing in savannas the Northern Territory of Australia and in
the frequently fire-ridden stated of Queensland. Quantification of the characteristics
3
of particle size distribution and emission factors is very important to understanding
the impact assessments of the fires in the states.
1.2. MOTIVATION OF THIS STUDY
As previously mentioned, the particles emitted during biomass burning have become
a serious global problem when you consider the huge areas burnt worldwide.
However, despite the scale of the problem and the impact of the associated particle
emissions on atmospheric processes, global change and human health, scientific
understanding of biomass burning processes remains poor. The reasons for initiating
this study were based on the following:
1. Knowledge of the characteristics of biomass burning particles (such as
particle size distribution and emission factors) is still limited.
2. The role of specific factors affecting the characteristics of biomass burning
particles is not clear.
3. PM2.5 emission data is very limited for many types of biomass burning, in
many countries around the world, particularly in Australia, a country that
frequently experiences biomass burning.
4. The information on particle number emission factors during biomass burning
is also very limited.
5. The characteristics of particles emitted during biomass burning in the
Northern Territory of Australia (whose fires significantly contribute to the
amount of particles in the atmosphere) are poorly understood. Moreover, such
characteristics are unknown for many of the states that experience biomass
burning every year in Australia.
4
6. The properties of biomass burning particles and the factors affecting biomass
burning particles during their transport in the atmosphere are poorly
understood.
7. The knowledge of vertical profiles of biomass burning particles in the
atmosphere is also very limited.
1.3. OVERALL OBJECTIVES OF THE STUDY The overall objectives in this study were:
1. To characterize biomass burning particle emissions in terms of particle size
distribution, particle number, and mass emission factors;
2. To investigate the factors influencing characteristics of biomass burning
particle emissions and quantify the magnitude of their impacts;
3. To characterize the vertical profile of biomass burning particles in the
atmosphere.
1.4. SPECIFIC AIMS OF THE STUDY The specific aims of the study were:
1. To design a system for characterizing biomass burning particle emissions in a
laboratory that will closely simulate real conditions in the field;
2. To investigate the impacts of burning rate on particle size distribution and
emission factors from biomass burning;
3. To characterize particle size distribution and emission factors during the
burning process (ignition, flaming, and smoldering);
4. To characterize particle size distribution and emission factors from the
burning of vegetation common to regions of Australia that are frequently
5
subject to fires, in particular the savannas of the Northern Territory of
Australia and Queensland;
5. To investigate the size distribution of biomass burning particles when
released from their sources and when retained in the atmosphere;
6. To determine a profile of particle concentrations at several different levels in
the atmosphere.
1.5. ACCOUNT OF SCIENTIFIC PROGRESS LINKING THE SCIENTIFIC
PAPERS
This thesis contains a collection of papers that have been published or submitted for
publication in refereed journals.
The study reported in the first paper (presented in Chapter 3) focused on
characterizing particle emissions from biomass burning in terms of size distribution
and emission factors. This study emphasized developing a system to characterize
biomass burning particles that would closely simulate real conditions in a field in
order to gain a better understanding of the size distribution and emission factors from
burning conducted under different environment conditions. The study aimed to
produce a controlled system that simulated biomass burning in a field that would
allow the investigation of a number of factors influencing particle emission
characteristics such as: species of tree burnt, part of tree (i.e., wood, leaves, and
branches), and burning rate. The study sought to quantify the particle emission
factors from combustion of trees typically found growing in South East Queensland
open forests under controlled laboratory conditions. A specific emphasis of the study
was on developing a better understanding of the size distribution and emission
6
factors from burning conducted under different environmental conditions. This study
presented a feasible simulation of the environmental conditions of a real forest fire
under controlled laboratory conditions. The results of this investigation has also
demonstrated that a number of factors, such as species of tree, part of tree, and
burning rate, influence particle size distribution, particle number emission factor, and
PM2.5 emission factor. The results showed that fast burning produces small particles
with high number emission factors and slow burning emits large particles with high
PM2.5 emission factors. Finally, the study quantified particle size distribution, particle
number emission factors, and PM2.5 emission factors from burning trees common to
South East Queensland forests. These results constitute an important contribution to
developing a quantitative assessment of the impact of fires in the state. The candidate
developed an experimental design and a scientific method, conducted experiments,
analyzed and interpreted data, and wrote the manuscript.
The second paper (Chapter 4) presents the results of study aimed in investigating
characteristics of particle emissions due to biomass burning in the Northern Territory
of Australia during fire seasons. The specific objectives of this study were to obtain
the characteristics of biomass burning particles during dry season and to investigate
the profile of biomass burning particles in different layers of the atmosphere by
measuring particle size distribution and particle number. In order to achieve the
goals, the airborne measurements were carried out along designated flight paths over
the savannas of the Northern Territory of Australia. The measurements were carried
out at several heights to characterize different atmospheric layers. The results
reported in this paper were based on data obtained during measurement campaigns
conducted in June and September 2003 by researches from the Queensland
University of Technology (QUT), the Defense Science and Technology Organisation
7
(DSTO), and the Australian Commonwealth Scientific and Industrial Research
Organization CSIRO. Specifically, the aerosol monitoring was conducted by Dr.
Zoran Ristovski (from 23rd to 27th of June 2003) and by Dr. Milan Jamriska (from
22nd to 26th of September 2003). The main focus of these activities was
characterization of biomass burning aerosols in the Northern Territory of Australia
during the dry season. The candidate had a role in processing, analyzing, interpreting
the data, and writing the manuscript. The results showed that biomass burning
particles emitted in the early dry season and late dry season had specific
characteristics, in terms of particle size distribution and particle number
concentration. The vertical profiles of biomass burning in the early dry season and
late dry season have also demonstrated significant differences. The results of this
study advance a scientific knowledge on the characteristics of ambient aerosol
emissions during fire seasons and particle profiles in the atmosphere. In addition, this
study contributes significantly towards advancing understanding of the effects of
biomass burning on atmospheric processes, as well as to modeling particle dispersion
in the atmosphere.
The third paper (presented in Chapter 5) is based on the results of both laboratory
and field studies on burnings a vegetation common to the Northern Territory of
Australia. The candidate organized the sample collection and transporting the
samples from the Northern Territory to Brisbane, Queensland; conducted laboratory
experiments; processed, analyzed and interpreted the laboratory and airborne
measurement data; and wrote the manuscript. The study characterized particle size
distribution from the burning of vegetation commonly found growing in the savannas
of the Northern Territory under controlled laboratory conditions. The size
distribution of particles obtained from the laboratory measurements was compared to
8
those from the airborne measurements to establish a comprehensive understanding of
the characteristics of the size distribution of particles from biomass burning in the
Northern Territory. The study demonstrated that a number of factors, including
variability of vegetation species growing in the area and burning conditions,
contributed to the characteristics of particle size distribution. The study also showed
that particle size distribution from biomass burning in the Northern Territory had a
specific characteristic for early dry season and late dry season. The vertical profile of
particles in the atmosphere during early dry season and late dry season also revealed
different characteristics. In general, the study has provided valuable information
regarding particle size distribution from biomass burning and several factors
influencing the distribution. The results of this study have contributed to expand
knowledge of the size distribution of biomass burning particles, particularly during
fire seasons in the Northern Territory of Australia, and have provided for a better
understanding of the characteristics of particle size distribution in general. The study
has provided important information on particle size distribution in the atmosphere,
which is necessary in order to obtain an estimate of the impacts of biomass burning
particles on atmospheric processes and human health. Overall, the results of these
studies advance a scientific knowledge on the physical processes of particles during
their transport in the atmosphere.
9
1.6. REFERENCES
ABS, (2004). Environment bushfires, Australian Bureau of Statistics, Accessed
March 2005, http://www.abs.gov.au/ausstats.
Andreae, M. O.,(1991), In Global Biomass Burning: Atmospheric Climatic and
Biospheric Implications; Levine, J.S.E., The MIT Press, Cambridge, MA.
Areskoug, H., P. Camner, S. E. Dahlén, L. Låstbom, F. Nyberg, G. E. Pershagen and
A. Sydbom, (2000), Particles in ambient air - a health risk assessment.
Scandinavian journal of work, Environment and Health, 26,1-96.
Bodhaine, B. A.,(1983), Aerosol measurements at four background sites. Journal of
Geophysical Research, 88, 10753 - 10768.
Cahoon, D. R., B. J. Stock, J. S. Levine, W. R. Cofer III and C. C. Chung,(1992),
Evaluation of a technique for satellite-derived estimation of biomass burning.
Journal of Geophysical Research, 97(D4), 3805-3814.
Dennis, A., M. Fraser, S. Anderson and D. Allen, (2002), Air pollutant emissions
associated with forest, grassland, and agricultural burning in Texas.
Atmospheric Environment, 36(23), 3779-3792.
Dockery, D. W., A. Pope, X. Xu, J. Spengler, D, J. H. Ware, M. E. Fay, B. G. Ferris
and F. E. Speizer, (1993), Mortality risk of air pollution: a prospective cohort
study. New England Journal of Medicine, 329, 1753-1759.
Etzel, R.,(1999), A research highlights: Air pollution and bronchitis symptoms in
Southern California children with asthma. Environmental Health
Perspectives, 107(9)
Ferge, T., J. Maguhn, K. Hafner, F. Muhlberger, M. Davidovic, R. Warnecke and R.
Zimmermann, (2005), On-line analysis of gas phase composition in the
10
combustion chamber and particle characteristics during combustion of wood
and waste in a small batch reactor. Environmental Science and Technology,
39(6),1393-1402.
Fine, P. M., G. R. Cass and B. R. T. Simoneit, (2002), Chemical characterization of
fine particle emissions from the fireplace combustion of woods grown in the
Southern United States. Environmental Science and Technology, 36, 1442-
1451.
Hays, M. D., C. D. Geron, K. J. Linna, N. D. Smith and J. J. Schauer, (2002),
Speciation of gas-phase and fine particle emissions from burning of foliar
fuels. Environmental Science and Technology, 36, 2281-2294.
Hedberg, E., A. Kristensson, M. Ohlsson, C. Johansson, P.-A. Johansson, E.
Swietlicki, V. Vesely, U. Wideqvist and R. Westerholm, (2002), Chemical
and physical characterization of emissions from birch wood combustion in a
wood stove. Atmospheric Environment, 36(30), 4823-4837.
Kaufman, Y. J., P. V. Hobbs, V. Kirchhoff, P. Artaxo, L. Remer, B. N. Holben, M.
D. King, D. E. Ward, E. M. Prins, K. M. Longo, L. F. Mattos, C. A. Nobre, J.
D. Spinhirne, J. Q. Thompson, A. M. Gleason, S. A. Christopher and S. C.
Tsay,(1998), Smoke, Clouds, and Radiation-Brazil (SCAR-B) experiment. J.
Geophys. Res-A, 103, 31783-31808.
Levy, J. I., J. K. Hammit and J. Spengler, D,(2000), Estimate the mortality impacts
of particulate matter: What can be learned from Between-Study Variability ?,
Enviromental Health Perspective, 108, 109-117.
Martins, J. V., P. Artaxo, C. Liousse, J. S. Reid, P. V. Hobbs and Y. J.
Kaufman,(1998), Effects of black carbon content, particle size, and mixing on
11
light absorption by aerosols from biomass burning in Brazil., J. Geophys.
Res-A,, 103, 32041-32050.
McConnel, R., K. Berhane, F. Gilliland, S. J. London, H. Vora, E. Avol, W. J.
Gauderman, H. G. Margolis, F. Lurmann, D. C. Thomas and J. M. Peters,
(1999), air pollution and bronchitis symptoms in Southern California
Children with asthma. Environmental Health Perspectives, 107, 757-760.
McDonald, J. D., B. Zielinska, E. M. Fujita, J. C. Sagebiel, J. C. Chow and J. G.
Watson, (2000), Fine particle and gaseous emission rates from residential
wood combustion. Environmental Science and Technology, 34, 2080-2091.
Nichol, J.,(1997), Bioclimatic impacts of the 1994 smoke haze event in southeast
Asia. Atmospheric Environment, 31(8), 1209-1219.
Ortiz de Zarate, I., A. Ezcurra, J. P. Lacaux and P. Van Dinh, (2000), Emission factor
estimates of cereal waste burning in Spain. Atmospheric Environment,
34(19), 3183-3193.
Peters, A., E. Liu, R. L. Verrier, J. Schwartz, D. R. Gold, M. Mittleman, J. Baliff, J.
A. Oh, G. Allen, K. Monahan and D. W. Dockery, (2000), Air pollution and
incidence of cardiac arrhythmia. Epidemiology, 11(1), 11-17.
Peters, A., S. Perz, A. Doring, J. Stieber, W. Koenig and H. E. Wichmann, (1999),
Increase in heart rate during an air pollution episode. American Journal of
Epidemiology, 150, 1094-1098.
Samet, J. M., F. Dominici, F. C. Curreiro, I. Coursac and S. L. Zeger, (2000a), Fine
particulate air pollution and mortality in 20 US cities. New England Journal
of Medicine, 343(24),1742-1749.
Samet, J. M., S. L. Zeger, F. Dominici, F. C. Curreiro, I. Coursac, D. W. Dockery, J.
Schwartz and A. Zanobetti (2000b). The national morbidity, mortality, and
12
air pollution study. Part II: Morbidity, Mortality and Air Pollution in the
United States, Health Effects Institute Research Report 94, Part II.
Schwartz, J., (1993), Particulate air pollutant and chronic respiratory disease.
Environmental Research, 62, 7-13.
Seiler, W. and P. J. Crutzen, (1980), Estimates of gross and net fluxes of carbon
between the biosphere and the atmosphere from biomass burning. Climatic
Change, 14,243-262.
Shaw, R. W.,(1987), Air pollution by particles. Science Environment, 255, 96 - 103.
Tolbert, P. E., J. A. Mulholland, D. D. MacIntosh, F. Xu, D. Danniel, O. J. Devine,
B. P. Carlin, M. Klein, J. Dorley, A. J. Butler, D. F. Nordenberg, H. Frunkin,
P. B. Ryan and M. C. White, (2000), Air quality and pediatric emergency
room visit for asthma in Atlanta, Georgia. American Journal of
Epidemiology, 151, 798-810.
Uherek, E.,2004. (Vegetation fire), Max Planck Institute for Chemistry, Mainz,
Accessed Mei 2004, http://www.atmosphere.mpg.de/enid/238.html.
WHO, (2000), Vegetation Fires. Http://www.who.int/mediacentre/factsheets/
fs254/en/print.html
Wieser, U. and C. k. Gaegauf, (2005),. Nanoparticle emissions of wood combustion
processes, Laboratories for Sustainable Energy System, Accessed March
2005,
http://www.oekozentrum.ch/downloads/publikationen/nanoparticles.pdf.
Wurzler, S. and M. Simmel, (2005), Impact of vegetation fires on composition and
circulation of the atmosphere, Accessed March 2005,
http://projects.tropos.de:8088/afo200g3/.
13
Yu, O., L. Sheppard, T. Lumley, J. Q. Koenig and G. G. Shapiro, (2000), Effects of
ambient air pollution on symptoms of asthma in Seattle- area children in the
CAMP study. Environmental Health Perspectives, 108(12), 1209-1214.
14
CHAPTER 2. LITERATURE REVIEW
2.1. Introduction This literature review aims at presenting a general overview of air quality, air
pollutants, and biomass burning. The general issues associated with Air Quality are
introduced in Chapter 2.2. The definition of biomass burning is briefly discussed in
Chapter 2.3. Physical and chemical characteristics of biomass burning are presented
in Chapter 2.4. Process measurements of biomass burning particles including
laboratory and airborne measurements are discussed in Chapter 2.5. Chapter 2.6
presents biomass burning in Australia. The impacts of biomass burning on human
health are discussed in Chapter 2.7. Dispersion models in general and for biomass
burning are presented in Chapter 2.8. Finally, the last chapters discuss the
knowledge gaps and conclusions of the reviews on particle characteristics from
biomass burning in general and particularly in Australia.
2.2. Air quality Quality of air has become a prime issue of importance for all countries around the
world. Air quality depends on particulate matter and gaseous pollutants produced by
a number of sources, including combustion, road dust, evaporation processes, and
others. The amount of pollutants in a volume of air, named pollutant concentration, is
associated with impacts on health and environment. Air pollutants in terms of either
particulate matter or gasses (carbon monoxide, sulphur dioxide, ozone, and nitrogen
dioxide), not only have a serious impact on human health but also play a role in
global change. The World Health Organisation (WHO) has set guidelines for a
number of pollutants to reduce their impacts on human health (WHO, 1999; 2006).
15
2.2.1. Airborne particle matter: general background and definitions
Particulate matter is a mixture containing many different components that come from
many sources. Particulate matter has compounds with local and regional variation.
Source, originality, and physical chemical properties determine the characteristics of
particulate matter (Englert, 2004). Different sources produce particulate matter with
a specific characteristic. Such sources include automobile and diesel trucks (Rogge et
al., 1993a; Schauer et al., 1999b), road dust and traffic debris (Rogge et al., 1993b),
steam boilers (Rogge et al., 1997), natural gas appliances (Rogge et al., 1993c),
natural vegetation emissions (Rogge et al., 1993a; Schauer et al., 2001),
broiling/cooking operations (Rogge et al., 1991; Nolte et al., 1999; Schauer et al.,
1999a), outdoor tobacco smoke (Rogge et al., 1994; Kavouras et al., 1998),
residential wood burning fire-places (Rogge et al., 1998; Prasad et al., 2001;
Kauffman et al., 2003) and biomass burning (Dennis et al., 2002; Hedberg et al.,
2002; Mukherji et al., 2002; Pagels, 2002; Reddy and Venkataraman, 2002a; Khalil
and Rasmussen, 2003; Wieser and Gaegauf, 2005; Wurzler and Simmel, 2005). The
contribution of each source to air pollution varies for different areas. In urban areas,
road transport contributes a large amount of particles. In general, biomass burning is
known as a major contributor of particulate matter to the atmosphere (Areskoug et
al., 2000; Ortiz de Zarate et al., 2000; Dennis et al., 2002; Rozenberg, 2002).
In terms of the physical properties of particulate matter, size is an important factor
not only in determining chemical composition of particles, but also in affecting
particle fate in the air. Smaller particles remain in the atmosphere longer than larger
particles. Small particles can remain airborne for days or weeks whereas large
16
particles are deposited within hours. Small particles can be transported for thousands
of kilometres while large particles are deposited closer to their sources. Size or
aerodynamic diameter of particulate matter is associated with particle measurements.
Large particles are measured based on their aerodynamic diameters (e.g. by filter),
however the aerodynamic diameters of very small particles with irregular shape are
unmeasurable. Measurements of the diameter are mostly based on the mobility of
particles. In terms of the impacts on human health, smaller particles can penetrate
deeper into the human respiratory system (WHO, 1999).
2.2.2. Particle size distribution
Pollutant sources naturally produce particulate matter with a variety of sizes called
polydisperse. The size of particulate matter can be characterised in terms of size
distribution. Particle size distribution is commonly represented using lognormal size
distribution, in which the particle concentration versus the particle size presented in
count median diameter (CMD) or mass median diameter (MMD) is Gaussian or
normal (bell shaped) when the particles are plotted on a logarithmic scale. Count
median diameter (CMD) is defined as a diameter for which half of the particles have
a smaller diameter and the other particles have a bigger diameter than the value.
While mass median diameter (MMD) is a diameter for which half the mass is
contributed by particles larger than the MMD and half by particles smaller than the
MMD (Baron and Willeke, 1993). The characteristics of the source are shown by the
lognormal size distribution and geometric standard deviation representing the width
of the peak of the distribution. Every different pollutant source may produce one or
more peaks of distribution called modes in every measurement. This means that the
source releases one or more particles of different size. Size distribution presented in
17
CMD can be useful to characterize the source signatures, for example the CMD of
some common vehicle emissions are about 0.1 µm for diesel vehicles, 0.05 µm for
spark ignition vehicles using unleaded petrol and 0.08 µm for LPG fuelled vehicles
(Morawska, 1998; Ristovski, 1998).
Airborne particles are found in a variety of sizes or modes. Particles of diameter
between 2.5 and 10 µm are called coarse particles. In terms of particulate matter
mass, they are called PM10. Coarse particles are mostly generated from mechanical
processes including grinding, breaking and wear of crystalline materials etc. Particles
having a diameter between 0.1 µm and 2.5 µm, also known as PM2.5, are named fine
particles. The major chemical constituents of fine particles are soot, condensated
acids, sulphates, nitrates, n-alkanes, polycyclic aromatic hydrocarbons (PAHs), n-
alkenoic acids, resin acids, and other toxins (Hays et al., 2002; Morawska and (Jim)
Zhang, 2002). Ultrafine particles are those with a diameter less than 0.1 µm and
largely consist of primary combustion products (from motor vehicles, biomass
burning, etc). A large proportion of ultrafine particles found in urban areas include
organic compounds, elementary elements, and metals arising from mobile source
emissions (Morawska, 1998; Kim et al., 2002). The majority of ambient particles are
recognized as ultrafine particles (Hinds, 1999). Formation of ultrafine particles in the
atmosphere has been attributed to at least three processes. The first is called direct
formation, describing particles that come from combustion processes associated with
traffic or industry sources (Kittelson, 1998), biomass burning (Reid et al., 2005;
Wieser and Gaegauf, 2005), or others that are emitted directly into the atmosphere.
Most ultrafine particles emitted by vehicle exhaust have a diameter in the range of
20-130 nm (Morawska, 1998) for diesel engines and 20-60 nm for gasoline engines
18
(Ristovski, 1998), and 30 to 200 nm for biomass burning (Hays et al., 2002; Hedberg
et al., 2002; Demirbas, 2004; Reid et al., 2005; Wieser and Gaegauf, 2005). The
second formation process is by nucleation and condensation from hot, supersaturated
vapour emitted when combustion is cooled to ambient temperatures (Stanier et al.,
2004). The last mechanism of ultrafine particle formation is chemical reactions in the
atmosphere that lead to the formation of low volatility species at ambient
temperature that may develop ultrafine particles through a variety of nucleation
processes (Kulmala et al., 2004; Stanier et al., 2004). A number of nucleation
mechanisms for ultrafine particles in the atmosphere have been proposed, such as
binary water-sulfuric acid nucleation (Kulmala and Laakssonen, 1990), ternary
water-sulfuric acid-ammonia nucleation (Kulmala et al., 2000), and ion induced
nucleation (Yu and Turco, 2000).
2.2.3. Emission factors
Sources produce pollutants in quantities presented using either emission factor or
emission rate (Hildemann et al., 1991a; Hildemann et al., 1991b). Emission factor is
defined as a unit based on a task (Mitra et al., 2002). For example, a task can be the
amount of energy generated by a power plant in g/Mj (Ahuja et al., 1987), the
amount of energy used for cooking stoves (Zhang et al., 2000), or the amount of
steam generated by a boiler in g/ton (Ge et al., 2001). It may be a task for a certain
distance driven by a motor vehicle that will be presented in g/km (Getler et al., 1998;
Winebrake and Deaton, 1999). Emission rate is defined as the amount of the
concerned pollutant as function of time. The unit of emission rate is expressed in g/s
or Tg/year. For biomass burning, emission factor is defined as the amount of
19
emission per kilogram of fuel burned or area burned (g/kg or Tons/ha): emission rate
is associated with the amount of emission per unit time (Tons/year).
2.2.4. Summary
Quality of air depends on the pollutant contained in the air. Sources determine the
kind and amount of pollutants released in ambient air. Sources produce particulate
matter with a variety of sizes. Size of particulate matter is important in determining
the characteristics of sources and in understanding the impact assessments. Sources
emit pollutants with a certain emission factor. Better knowledge of pollutants in
terms of definition, kind, characteristic, sources, and emission factor is necessary to
get a better understanding of air quality in general.
2.2.5. Concentration of particulate matter in different countries
The concentrations of particulate matter, such as coarse particles PM10, fine particles
PM2.5, and ultrafine particles, in ambient air may be associated with quality of air and
also related to the impacts either on human health or the environment. The
concentrations of particulate matter vary for different countries. Figure 1 presents an
example of the concentration of PM2.5 in several cities in the world. It can be seen
that the concentration of particulate matter PM2.5 is 7.6 µg/m3 for Brisbane, Australia
(Thomas and Morawska, 2002); 67.6 µg/m3 for Shanghai, China (Ye, 2003); 22.4
µg/m3 for Taen, South Korea (He, 2003); 16.1 µg/m3 for Ho Chi Min, Vietnam
(Hien, 2001) and 56.9 µg/m3 for Jakarta, Indonesia (Zou, 1997). Shanghai and
Jakarta are the cities with the highest concentration of PM2.5.
20
0
20
40
60
80
ShanghaiChina
TaeanSouthKorea
Ho ChiMinh
Vietnam
BrisbaneAustralia
JakartaIndonesia
City / Country
Conc
entra
tion
(µg/
m3)
Figure 2.1.The concentration of PM2.5 in several cities around the world.
Figure 2.2. Concentration of PM2.5 and PM10 in several cities of Australia (Ayers et
al., 1999)
Figure 2.2. presents another example of PM2.5 and PM10 concentrations measured at
six different cities in the Eastern regions of Australia during 1996 and 1997 (Ayers et
al., 1999). Figure 2 shows that the highest level of PM2.5 and PM10 is in Launceston
with concentrations of 37.5 µg/m3 for PM2.5 and 49 µg/m3 for PM10 due to wood
21
burning for heating. The lowest level of PM2.5 and PM10 is found in Brisbane with
concentrations of 5.7 µg/m3 for PM2.5 and 19.1 µg/m3 for PM10. The concentration
of PM2.5 in Brisbane shown in Figure 2 is comparable with those measured in other
studies: 7.3 µg/m3 (Chan et al., 1999) and of 7.6 µg/m3 (Thomas and Morawska,
2002). The concentrations of PM2.5 and PM10 in Melbourne seen in Figure 2 are
much less than those found by another study for six different suburbs in Melbourne,
where the average concentrations of PM2.5 and PM10 were 34 µg/m3 and 50.3 µg/m3
respectively (Hurley et al., 2003).
2.2.6. Summary
The concentration of particulate matter varies for different cities or countries in the
world. The data show that the concentration of particulate matter in developing
countries is relatively higher than that in developed countries. However in a number
of cities in developed countries, the concentration of particulate matter is found to be
high. This implies that air pollutants in terms of particulate matter have become a
problem for every country in different respects.
2.2.7. Ambient air quality standards
A concentration of particulate matter in ambient air has been standardized. Different
countries have their own standards for particulate matter concentration in ambient air
depending on the policies of the country. Table 2.1 shows the ambient air quality
guidelines for particulate matter in several countries.
22
Table 2.1. Ambient air quality guidelines for particulate matter
Country Particulate matter
Concentration (µg/m3)
Averaging period
Date of implementation
Ref
150 24 hours 17 December 2006
a PM10
Revoked1 Annual1 a 35 24 hours 17 December
2006 a USA
PM2.5 15 Annual 17 December 2006
a
50 24 hours 31 December 2004
b UK PM10
40 Annual b 50 24 hours 1 January 2005 c EU PM10 40 Annual c
PM10 50 24 hours June 1998 d 25 24 hours 2005 d Australia PM2.5 8 Annual 2005 d 50 24 hours May 2002 e New
Zealand PM10 20 Annual May 2002 e 200 1 hour 8 May 1973 f Japan PM10 100 24 hours 8 May 1973 f 50(i),150(ii),250(iii) 24 hours January 1996 g China2 PM10
40(i),100(ii),150(iii) Annual January 1996 g (1) Due to a lack of evidence linking health problems to long-term exposure to coarse particle pollution, the agency revoked the annual PM10 standard in 2006 (effective December 17, 2006). (2) (i) Sensitive areas of special protection; (ii) typical urban and rural areas and (iii) special industrial areas.
a. http://www.epa.gov/air/creteria.html b. http://defra.gov.uk/environment/airquality/airqual/index.html c. European commission Guidelines Website d. http://www.ephc.gov.au/nepms/air/air_nepm.html e. http://www.mfe.govt.nz/publications/air/ambient_guide_may02.pdf f. http://www.env.go.jp/en/lar/regulation/aq.html g. http://www.cepis.ops-oms.org/bvsci/e/fulltext/normas/normas.html
The US government through the US Environmental Protection Agency standardizes
the concentration of PM10 at 150 µg/m3 in average over 24 hours, which is not
exceeded more than once per year on average over 3 years. The agency revoked the
23
annual PM10 standard of 50 µg/m3 effectively from 17 December 2006. The ambient
air quality standard for PM2.5 concentration is set at 35 µg/m3 for a 24 hour averaging
period and 15 µg/m3 annually. The United Kingdom and Europe set the average
PM10 concentration at 50 µg/m3 for 24 hours and 40 µg/m3 annually as the ambient
air quality standard. Japan has a standard for ambient air quality of 100 µg/m3 for
PM10 which was established in 1973, just prior to problems with air pollution caused
by particulate matter and relying on the emergence of recent evidence of health
effects. Australia, through the National Environment Protection Council, agreed to
set uniform standards for ambient air quality in June 1998. The council commenced
applying the standard of PM2.5 of 25 µg/m3 per day and 8 µg/m3 per year in 2005.
The maximum level of PM10 was set at 50 µg/m3 on average for a day. New Zealand
regulations require the average maximum concentration of PM10 not exceed 50
µg/m3 for 24 hours and 20 µg/m3 annually. China standardizes the maximum level of
PM10 depending on areas. The World Health Organization set guidelines for global
air quality for PM10 in a graph presenting exposure-response slopes with no threshold
(WHO, 1999). In 2006, the WHO released new air quality guidelines with
dramatically lower standards for levels of pollutants by reducing particulate matter
pollution PM10 from 70 to 20 µg/m3 for annual mean levels, and PM2.5 from 35 µg/m3
to 10 µg/m3 for annual mean levels. The new standard is expected to reduce deaths in
polluted cities by 15 % every year (WHO, 2006).
2.2.8. Summary
Standards or guidelines for air quality in ambient air has been set in most countries to
maintain the quality of air, keeping pollutant concentrations under threshold levels to
24
limit their impacts on human health. Every country has ambient air quality standards
depending on the conditions of the country. WHO has set air quality guidelines to
provide uniform targets for air quality addressed to all countries around the world.
2.2.9. Concentration of gaseous pollutants in different countries
Several gaseous pollutants have been recognized as having serious effects on human
health and global change, including carbon monoxide (CO), carbon dioxide (CO2),
ozone (O3), nitrogen dioxide (NO2), sulphur dioxide (SO2), ammonia (NH3), volatile
and semi-volatile organic compounds (VOCs and SVOCs). Below is a brief review
of studies on the level of gaseous pollutants in different countries.
The concentrations of O3, SO2, and NO2 were measured for large cities in developed
and developing countries (Baldasano et al., 2003). It has been found that the
concentrations of the gaseous pollutants are significantly higher in the developing
countries than in developed countries. Carmichael et. al (2003) measured SO2, O3,
and NH3 concentrations in Asia, Africa, and South America. They found that Linan,
China had the highest SO2 concentration of 13.06 ppb. Zoetele (Cameroon), Isla
Redonda and Ushuaia (Argentina) were the cities with the lowest concentrations of
SO2, which were less then 0.03 ppb. The highest concentration of NH3, which was
more than 40 ppb, was found in Agra, India. Ozone was found in the high
concentration in China, Taiwan, India and Turkey with concentrations of more than
30 ppb.
25
2.3. Biomass Burning: General Background
2.3.1. Definitions
Biomass
Biomass is defined widely as vegetations or plants in which the plant tissues contain
structural and non structural carbohydrate components as products of photosynthesis
process, such as cellulose, hemicelluloses, lignin, lipid, proteins, simple sugars,
starches, water, hydrocarbon components (HC), ash, and other compounds (Jenkins
et al., 1998; Simoneit, 2002). A variety of the compounds in biomass depend on
species, places of growth, condition of growth, and type of plant tissue. Cellulose is a
major compound of biomass containing a long-chain linear polymer built of 7000–
12,000 D-glucose monomers and individual cellulose molecules linked with
glycosidic bonds, which is responsible for about 30 % of structuring of plant tissues
(Milne, 1990; Jenkins et al., 1996a). Hemicelluloses are polysaccharides consisting
of carbon monosaccharide units: glucose, mannose, galactose, xylose, and arabinose.
Hemicellulose molecules are also polymers of 100–200 monomers, containing fewer
sugar monomers than cellulose molecules (Parham, 1984; Simoneit, 2002). Lignin is
an irregular biopolymer of phenylpropane units derived from p-coumaryl, coniferyl
and sinapyl alcohols (Schultz, 1989) and contains anisyl, vanillyl (guaiacyl) and
syringyl nuclei (Simoneit, 1993; Rogge et al., 1998). Cellulose, hemicellulose, and
lignin, play an important role in emission production from biomass burning.
Biomass is highly oxygenated because 30–40 % of dry biomass contains oxygen,
which is the primary component in the burning process. The major constituent of
biomass is carbon, which is 30–60 % of dry biomass depending on ash content.
26
Other components such as nitrogen, sulphur, and chlorine are less than 1 % of dry
matter, which contribute to pollutants released from biomass burning (Miles, 1995).
Inorganic compounds are also found in biomass, including potassium (1 %), dry
matter, silica (10– 5 %) and dry matter (Jenkins et al., 1998). The concentration and
composition of the elements in biomass determine the properties of biomass burning.
Burning process
Burning, or combustion, is a complex process involving chemical and physical
reactions, and transfer of mass and heat. Burning is also defined as a combination of
reactants such as fuels, water, and air; reacting globally to produce some products of
burning or emissions (Jenkins et al., 1998). Every reactant fuel produces different
characteristics of burning. At least 15 elementary compounds, such as C, H, O, N, S,
Cl, Si, K, Ca, Mg, Na, P, Fe, Al, and Ti; that compose the fuels in certain ratios will
determine the characteristics of the burning process. The moisture content of fuels
also determines the burning process. If the moisture content of fuels is high, fuel
does not spontaneously react and some amount of energy is needed to evaporate the
water. This reduces the heating value of the fuels and decreases the efficiency of the
burning. These phenomena were observed in a study of wood burning (Core, 1982;
1984; Jenkins et al., 1998). On the other hand, less moisture content causes the fuel
to burn faster leading to incomplete burning, which increases smoke particle
formation. The supply of oxygen during burning has been recognized as a significant
factor contributing to emission production (Zou et al, 2003).
27
Burning is also defined as a process involving hydrolization, oxidation, dehydrating,
and pyrolyzation in increasing of temperature (Simoneit, 2002). The burning process
is divided into ignition, flaming, and smouldering phases. Ignition known as the
starting process of burning is a process to oxidize fuels in raising temperature.
Flaming involves hydrolization, oxidation and dehydrating processes. Flaming is
commonly known as the real burning phase. During this phase, the released heat
provides the energy that is necessary to evaporate the water in the cells of the
biomass (Skaar, 1984) and to decompose the compounds of the biomass, including
cellulose, hemicellulose and lignin; to become monomers (Hornig, 1985). In this
phase, moisture content of the biomass determines the completeness of the biomass
burning. High moisture content biomass means more energy is needed to vaporize
the water, decreases the efficiency of burning, and increasing smoke formation
(Core, 1984). On the other hand, low moisture content biomass burns fast, causing
oxygen-limited conditions and leading to incomplete burning and increasing smoke
formation. Smouldering is a phase in which there is enough heat to oxidate the
reactive char (solid phase of burning) and to decompose the biomass into some
products.
Biomass Burning Process
From the definitions above, biomass burning is described as a complex process to
decompose biomass compounds such as cellulose, hemicelluloses, and lignin; by
increasing their temperature. At temperatures below 300 oC, cellulose overcomes
depolymerisation, dehydration, fragmentation, and finally oxidation to lead to char
formation. At temperatures greater than 300 oC, cellulose overcomes bond-splitting
28
by transglycosylation, fission, and disproportionation reactions into anhydro sugar
and volatile products, such as levoglucosan (1,6- anhydride of glucose), furanose
isomer, and other anhydrides (Shafizadeh, 1984; Hornig, 1985; Simoneit, 2002). At
this stage, lignin is degraded into monomers such as coumaryl, vanillyl, and syringyl
moieties (Simoneit, 1993). Levoglucosan was found in fine particles. It was used to
trace cellulose in biomass burning (Hornig, 1985; Simoneit, 1999; Oros and
Simoneit, 2001). Most of the derived compounds of cellulose and lignin were used as
tracers for biomass burning (Oros and Simoneit, 2001) and detected in the
atmosphere (Rogge et al., 1993d; 1998; Simoneit and Ellias, 2001b).
2.3.2. Summary
Burning of biomass involves complex processes to break apart the biomass
molecules and decompose compounds into burning products or emissions. The
transformation of biomass compounds during the burning process that determines the
characteristics of the subsequent emissions has been poorly understood due to the
complexity of the mechanisms involving physical and chemical processes. Several
factors such as those related to the biomass itself, the burning process, or others, may
contribute and play a role in the mechanisms of biomass burning. To have a better
understand of biomass burning, it is necessary to investigate complex issues,
including the burning process, transformation of biomass compounds into other
products, the characteristics of products in every phase of burning, factors
influencing the characteristic of biomass burning and emission production, and the
relationships between these factors.
29
2.3.3. Areas of biomass burning
Biomass burning, including controlled and uncontrolled forest and savannah fires
and agricultural burning, has consumed huge areas around the world. Biomass
burning in Texas, including wildfire, prescribed wild land, prescribed line,
agricultural, logging slash and land clearing slash fire destroyed areas of 0.5 million
hectares in 1996 and 1997 (Dennis et al., 2002). More than 1.3 million ha of forest
were burnt in China in 1987. In the same year, forest fires in eastern Asia consumed
approximately 14 million hectares (Cahoon et al., 1992). Forest burning destroyed
more than 5-20 million hectares in Indonesia in 1994 and 1997 (Nichol, 1997). Other
data show that 10 million hectares of forest in northern latitudes; 40 million hectares
of tropical and sub tropical forest, and 500-1000 million hectares of open forest and
savannahs are burnt every year (Uherek, 2004). More than 5 million hectares of
forest was destroyed by fires in Kalimantan during the 1982-83 El Nino drought, and
9 million hectares of vegetation were burnt in Sumatra and Kalimantan in 1997-98.
Between 5 and 20 million hectares of forest are consumed by uncontrolled fire every
year in North America and Eurasia. In Australia, 40 – 130 million hectares of land
are burned annually (WHO, 2000).
2.3.4. Summary
Significant vegetation areas have been burned around the world. Several areas
experience biomass burning every year. This shows that biomass burning is a serious
problem. A large amount of emissions has been released into the atmosphere as a
consequence of the burning.
30
2.3.5. Biomass burning emissions
2.3.5.1. Emission rate
Biomass burning has been recognized as a major contributor of particles and gases to
the atmosphere at a variety of rates. Biomass burning in Texas in 1996 emitted
particulate matter of 161,000 tons/year, and gaseous compounds including CO, CH4,
NOx, NH3 and NMHC of 698,000 tons/year (Dennis et al., 2002). Wood burning in
Sweden produces particulate matter of 8,600 – 65,000 tons/year, almost half of the
total emissions of the world which is 49,000 – 112,000 tons/year (Areskoug et al.,
2000). The total emission produced by burning of agricultural wastes is predicted in
the range of 1700 to 2100 Tg of dry matter (dm) per year (Seiler and Crutzen, 1980).
Burning of cereal wastes in Spain releases total particulate matter (TPM) of 80 – 130
Gg, NOx of 17 – 28 Gg, CO of 210 – 350 Gg, and CO2 of 8 – 14 Gg annually (Ortiz
de Zarate et al., 2000).
2.3.5.2. Type of emissions
Emissions of biomass burning consist of a wide range of particles and gasses that
affect atmospheric processes and human health. The emission of CO, CH4 and
volatile organic compounds (VOC) affect the oxidation capacity of the troposphere
by reacting with OH radicals, nitric oxide (NO) and VOC to lead to the formation of
ozone and other photo oxidants (Koppmann et al., 2005). The World Health
Organization (WHO) identifies a number of emissions of biomass burning that affect
human health, and places them into a number of classes: particulate matter,
polynuclear/polycyclic aromatic hydrocarbons (PAH), Carbon monoxide (CO),
aldehydes, organic acids, semi-volatile and volatile organic compounds, nitrogen and
31
sulphur-based compounds, ozone and photochemical oxidants, inorganic fraction of
particles, and free radicals.
Particulate matter released from biomass burning was measured from burning of
different biomass, such as woods (Hedberg et al., 2002; Khalil and Rasmussen, 2002;
Mukherji et al., 2002; Pagels, 2002; Reddy and Venkataraman, 2002a), grassland
(Dennis et al., 2002), agriculture (Anderson, 2002; Dennis et al., 2002; Reddy and
Venkataraman, 2002a)and dung cake and biofuel briquette (Mukherji et al., 2002;
Reddy and Venkataraman, 2002a). The majority of particles resulting from biomass
burning were reported to be less than 2.5 μm in diameter (Hueglin et al., 1997;
Hedberg et al., 2002; Ferge et al., 2005; Wieser and Gaegauf, 2005).
Polynuclear/polycyclic aromatic hydrocarbons (PAHs) are known as carcinogenic
particles contributing to huge affects on human health, especially in the development
of cancer in the human body (Harvey, 1991; WHO, 1999). The mechanisms of PAH
formation during burning of organic matter is not fully understood, particularly the
basics of formation of PAH that happens when the radicals produced by burning at
high temperatures recombine to become PAH at lower temperatures (Simoneit, 1998;
Simoneit, 2002). PAHs including acenaphtene, acenaphthylene, anthracene,
benzo[a]pyrene, benz[a]anthracene, dibenz[a.h]anthracne, fluoranthene, naphthalene,
phenanthrene, and pyrene have been identified as having serious effects on human
health (WHO, 1999; Morawska and (Jim) Zhang, 2002). Benzo[a]pyrene is one of
the PAHs contributing to cancer development in human cells. PAH compounds were
32
found in wood burning (Oanh et al., 1999; Oros and Simoneit, 2001; Hedberg et al.,
2002; Zou et al., 2003). Oros and Simoneit, 2001, detected more than 30 PAHs
emitted by burning of different kinds of wood. Several of these PAHs were identified
as mutagenic and having genetoxic potential, such as nez[a]anthracene,
benzo[a]pyrene, and cyclopenta[c,d]perene (Arcos and Argus, 1975; IARC, 1989).
Herberg et al, 2002, reported that fluorene, phenanthrene, anthracene, fluranthene,
and pyrene contributed to more than 70 % of the mass of PAHs for birch wood
burning. Another study measuring PAH emissions of wood burning found PAH and
genotoxic PAH levels of 11,508 µg/kg and 953 µg/kg respectively (Zou et al., 2003).
A previous study found 110,200 µg/kg for the total PAHs and 13,400 µg/kg for
genotoxic PAHs (Oanh et al., 1999). Characteristics of PAH emission from biomass
burning were investigated as a factor of type of wood and combustion appliances
(McDonald et al., 2000), and moisture (Korenaga et al., 2001). The results showed
that these factors influenced PAH emissions from biomass burning.
Carbon monoxide (CO) is a colourless and odourless toxic gas produced by the
incomplete burning of biomass, such as wood burning (Muraleedharan et al. 2000;
Osán et al. 2002; Dennis et al. 2002; Prasad, V.K. 2000, Reddy and Venkataraman
2002; Edward et al. 2003) and agricultural and grassland burnings (Dennis et al.
2002). Carbon monoxide is a major compound produced by biomass burning, with
emission factors of about 130 g/kg wood burned (EPA, 1986; Larson and Koenig,
1993). Muraleedharan and colleagues reported that the carbon monoxide emission
factors from peat combustion was 37 g/kg (Muraleedharan et al., 2000). Burning of
cereal emitted carbon monoxide with an emission factor of 35 g/kg per kilogram
33
(Ortiz de Zarate et al., 2000). Carbon monoxide coming from biomass burning was
reported to contribute 32 % of the total of carbon monoxide produced from other
sources (Levine, 1990).
Aldehydes are chemical compounds also recognized as toxic gases that are extremely
irritating and cause respiratory problems. A numbers of aldehydes were found as
products of wood burning, including formaldehydes, acetaldehyde, crotonaldehyde,
benzaldehyde, isovaleraldehyde, tolualdehdes (Hedberg et al., 2002). Schauer and
co-workers (2001) reported emission factors for aldehydes resulting from wood
burning of 4.1 g/kg for pine, 2.1 g/kg for oak, and 2.7 g/kg for eucalyptus. The major
aldehydes found in this case were acetaldehyde and formaldehyde. Pine produced
more acetaldehyde and formaldehyde than oak and eucalyptus (Schauer, 2001).
Organic Acids are produced by biomass burning. The organic acid (3,4,5)-
trimethosybenzoic (TMBA) was found in emissions from the burning of birch wood
(Hedberg et al., 2002) and oak wood (Simoneit, 1993), with emission factors of 26
mg/kg and 23 mg/kg respectively. Other types of organic acids were also emitted by
the burning of oak and eucalyptus.
Semi-volatile and volatile organic compounds (SVOCs and VOCs) were found in
biomass burning emissions. Herberg et al, 2002, measured a number of VOC
emissions from birch wood burning, including toluene, benzene, and acetone; with
emission factors of 740 mg/kg, 1500 mg/kg, and 366 mg/kg respectively. Schauer
(2001) measured the emission factors of VOCs for several types of wood burning.
34
Benzene and toluene were reported with emission factors of 383 mg/kg and 158
mg/kg respectively. Acetone was found with a variety of emission factors depending
on the type of wood: 462 mg/kg for oak, 749 mg/kg for pine, and 79 mg/kg for
eucalyptus (Schauer, 2001). Another study of VOCs from wood burning was
conducted by McDonald et al, using different combustion appliance. The study
reported that the emission factors of ketones, benzene and toluene released from hard
wood were greater than those measured for soft wood (McDonald et al., 2000).
Monitoring of biomass burning gases in South-east Asia and India found VOCs such
as methanol, acetone, acetonitrile, isopere and methyl vinyl ketone (MVK), and
methacrolein (MACR) in significantly high quantities (Karl et al., 2003).
Nitrogen and sulphur-based compounds. Biomass burning has been recognized as
producing nitrogen and sulphur-based compounds (Prasad 2000; Korenaga 2001;
Zarate et al. 2002; Osán et al. 2002; Reddy and Venkataraman 2002; Dennis et al.
2002; Anderson et al. 2002; Khalil and Rasmussen, 2003). Ammonia gas is one of
the compounds found in several kinds of biomass burning. Ammonia gas is known
to have a relatively short lifetime in the atmosphere of a few hours to a few days
(Warneck, 1988; Dentener and Crutzen, 1994). In contrast, ammonium ions as an
aerosol categorized in PM2.5 have a lifetime in the order of 1 – 15 days (Aneja et al.,
2001).
Inorganic elements such as Cu, Fe, Pb, Mn, Zn, Al, Mg, Si, Ca, Ti, Mn, Ni, Po, Cd,
Na, Cl, S, K and V were commonly found from biomass burning (Osan et al., 2002;
Khalil and Rasmussen, 2003). Osan and colleagues measured inorganic elements
35
from wood burning and found most of these elements in the diameter range of coarse
particles and fine particles (Osan et al., 2002). A study of wood burning that was
conducted by Hays et al, reported that the emission factors of inorganic compounds
depended on the kind of wood burned (Hays et al., 2002). They reported that most of
the inorganic particles had diameters of less than 2.5 μm.
2.3.6. Summary
Significant amounts of emissions have been released by biomass burning into the
atmosphere every year. The emissions have been recognized as a major contributor
of particulate matter and gasses to the atmosphere. Most of the emissions have been
known to have serious effects on human health and atmospheric processes.
2.4. Particles originating from biomass burning
2.4.1. Particle formation
Attempts have been made to understand particle formation in biomass burning for
many years (Pyne, 1984; Lobert and Warnatz, 1996; Simoneit, 2002). However due
to the many factors influencing particle formation and complexity during burning
processes, knowledge of particle formation is very limited. Reported data in literature
state that particle formation in the burning process is started by the creation of
condensation nuclei such as polycyclic aromatic hydrocarbons (PAHs) from ejected
fuel gases (Frenklach 2002) and other species, including alkanes (Kent 1986; Turns
1996), in flames. The PAH molecules overcome chemical and coagulative processes
to grow to between 3000 and 10000 atomic mass units and become condensation
nuclei for other pyrolized species and may experience growth (Reid et al, 2005).
36
Many of the species may then reduce in size through further oxidation in the flame
zone if temperatures exceed 1100 K. If the temperature is insufficient to complete
oxidation (T < 1100 K), the particle size may increase to a secondary condensation
growth (Glasmman 1977). Current knowledge is not sufficient to fully understand
particle formation in biomass burning, especially related to burning phases, the role
of oxygen in the process and the physical and chemical processes in forming of
particles. The process of particle formation influences the characteristics of particle
emissions.
2.4.2. Particle composition
Biomass burning particles consist of two large components: organic carbons such as
VOCs, PAHs and black carbon; and inorganic elements. Organic carbon and black
carbon compose 50 to 70 % of the mass of all particle emissions, and about 55 % and
8 % of fine particle mass is classified as these compounds respectively. The ratio of
black carbon to organic carbon in biomass burning particles varies in the range of 1:8
to 1:12. (Cachier et al., 1995; Liousse et al., 1995; Andreae et al., 1998; Ferek et al.,
1998; Formenti et al., 2003). Black carbon concentrations emitted during biomass
burning vary according to burning phase. The smouldering phase produces black
carbon with concentrations in the range of 2-27 %. In contrast, higher concentrations,
with variation by a factor of over 5, is emitted during flaming phase (Reid et al.,
2005). Natural conditions contribute to the variation of black carbon concentrations.
Black carbon content has been found to vary from 2-30 % in tropical forest (Ferek et
al., 1998; Reid and Hobbs, 1998), and 4-28 % in savannah (Cachier et al., 1995;
Liousse et al., 1995; Andreae et al., 1998; Formenti et al., 2003). Other components
37
of biomass burning particles are inorganic elements. Approximately 10 % of fine
particles from fresh smoke is composed of inorganic compounds such as potassium,
chlorine, and calcium (Andreae et al., 1998; Ferek et al., 1998). Inorganic elements
play a role in a significant enrichment of species associated with secondary aerosol
production.(Reid et al., 2005).
2.4.3. Summary
Particle formation in biomass burning has been poorly understood due to complexity
of processes and factors influencing their formation. Transformation of biomass
compounds by increasing temperature, particle formation in burning phases, and the
role of oxygen on particle formation during burning are complex factors contributing
to the lack of knowledge of particle formation in biomass burning. The available
information presented in literature is still insufficient for full understanding of
particle formation in biomass burning. In real fires, there are other parameters such
as fuel variability, fuel density, and burning environment further adding to the
complexity of particle formation.
2.4.4. Characteristics of particle size
Biomass burning emits particles with a large variety of sizes. Particle size
distribution from fresh biomass burning smoke has been reported in the range of 30
to 500 nm depending on the type of vegetation, its moisture content, and the burning
system. Measurements of particle emissions from wood burnt in several types of
wood combustion systems found particles with diameters in the range of 30 nm to
300 nm (Wieser and Gaegauf, 2005). Herberg et al. in 2002 reported that particles
38
produced by the burning of Birch wood were found to have diameters in the range of
20 nm to 500 nm. A study conducted by Hueglin and his colleagues measuring the
size distribution of particles emitted by Beech wood with moisture content between
15 – 18 % using a residential wood stove obtained particle diameters of 270 nm in
the beginning of the burning process, of 170 nm during the flaming process, and
about 50 nm to 60 nm towards the end of the burning process (Hueglin et al., 1997).
Kleeman and associates found particles to range in diameter from 100 nm to 200 nm
(Kleeman et al., 1999).
Characteristics of biomass burning particles are dependant on the place of biomass
growth. Hays and colleagues reported that open burning of the foliar fuels collected
from native US habitats produced particles with diameters between 150 nm and 200
nm for ignition, 100 nm and 150 nm for flaming, and 70 nm and 150 nm for
smouldering; depending on the species of wood (Hays et al., 2002). A study
conducted by Hueglin and his colleagues, using a residential wood stove for burning
Beech woods taken from Sweden with moisture content between 15 – 18 %, found
that most of particles had diameters centred at 170 nm during the flaming process,
and 60 nm at the end of the burning process (Hueglin et al., 1997). A study of
burning a similar species of wood reported diameters of particles in the range of 30
nm to 130 nm (Hedberg et al., 2002).
Field measurements show that the size distribution of burning biomass particles
depends on region. The measurements of particle size distributions in African
savannah (Anderson, B. E. et al., 1996) and South Africa (Le Canut et al, 1996)
39
found most particles with count median diameters (CMD) of (220 ± 3) nm and (180
± 60) respectively. Measurements of biomass burning particles in North America
obtained particles with diameters of (190 ± 3) nm (Hobbs, P.V et al., 1996). Particles
with diameters in the range of 120 to 230 nm were measured from aged smokes in
Brazil (Anderson, B.E et al., 1996; Reid et al., 1998b). Particles released from
biomass burning in the Amazonia were reported with diameters in the ranges of 15 to
279 nm (Guyon et al., 2005a) and 51 to 144 nm (Krejci et al., 2005). The differences
in particle size distribution of biomass burning may be due to fire intensity, fuel
density, and intense regional burning. For example, in the tropics numerous fires
constantly occur and continuously eject fresh smoke into the air.
Characteristics of particle size distribution from biomass burning in the atmosphere
are very complex due to several atmospheric processes, including physical, chemical,
and thermodynamical processes. Understanding the characteristics involves
complicated studies of physical properties, chemical properties, and thermodynamic
properties of particles. In terms of physical properties, biomass burning particles
undergo growth during transport in the atmosphere. Particles in the atmosphere have
been detected to increase in size with age of smoke (Reid et al., 1999). Most aged
particles are known as secondary particles (Andreae et al., 1998; Reid et al., 1998a;
Formenti et al., 2003; Gao et al., 2003). Growth rate of biomass burning particles
varies from 1 hour to a time scale of days after emission (Radke et al., 1995; Hobbs,
P. V et al., 1996; Reid et al., 1998a; Abel et al., 2003). Limited data regarding the
factors playing a role in particle growth in the atmosphere causes the lack of
understanding of characteristics of biomass burning particles. Chemical and
40
thermodynamic processes of particles in the atmosphere have been also poorly
understood.
2.4.5. Summary
The characteristics of the size distribution of biomass burning particles have been
reported in several studies. Particle size distribution of biomass burning shows a
variety depending on type of biomass, moisture contents of fuels, and way of
burning. The relationships among the factors, including burning processes or burning
phases, particle formation, and rate of burning, as they relate to the characteristics of
biomass burning particles have been poorly understood. A variety of particle size
distribution from biomass burning was found from field measurements in which the
size of particles depended on fuel variability and burning environment that differ in
different places. However, the relationships between those factors and burning
processes in the field are very complicated. The current limited knowledge of these is
insufficient for understanding the characteristics of biomass burning.
2.4.6. Particle emission factors
Emission factors, measured in mass or number of particles produced per unit mass
biomass fuel burned (g/kg and number/kg), are important in understanding the
impacts of biomass burning on human health, global and climate change or for
modelling smoke particle production into the atmosphere. In terms of human
impacts, emission factor is associated with dose of particle emissions received by
humans, and scope of areas covered by particle emissions. The emission factor of
particles is also related to the amount of particles emitted in the atmosphere that
41
affects directly the radiation balance and indirectly the acidification of clouds, rain,
and fog.
Concerning the impacts of biomass burning, previous studies have been focused on
measurements of the emission factor of particulate matter PM2.5. There have been
several studies reporting PM2.5 emission factors for different species of trees in the
relevant literature. A study of an open burning of mixed hardwood forest foliage in
the US, found that the PM2.5 emission factor was 10.8 ± 3.9 g/kg (Hays et al., 2002).
The PM2.5 emission factors from burning of woods grown in the North-eastern
United States were measured in the range of 2.7 to 5.7 g/kg for hard woods and 3.7
to 11.4 g/kg for soft woods (Fine et al., 2001), while similar study of the woods
grown in the Southern United States yielded emission factors in the range of 3.3 to
6.8 g/kg for hard woods and 1.6 to 3.7 g/kg for soft woods (Fine et al., 2002). A
study aimed at characterization of emissions from wood burning in a fireplace found
that the emission factors were 2.9 to 9 g/kg for softwoods and 2.3 to 8.3 g/kg for
hardwoods (McDonald et al., 2000). The burning of Birch wood in a stove produced
particles PM2.5 with emission factors of 0.1 to 2.6 g/kg (Hedberg et al., 2002). The
PM2.5 emission factors from the burning of wood logs in several combustion systems
were reported in the range of 0.13 to 1.68 g/kg (Wieser and Gaegauf, 2005). Figure 5
presents PM2.5 emission factors from wood burning measured by previous studies.
42
0
2
4
6
8
10
12
14
16
Hays,
2002
Fine, 2
001
Fine, 2
001
Fine, 2
002
Fine, 2
002
McDon
ald, 2
000
McDon
ald, 2
000
Hedbe
rg,20
02
Wies
er an
d Gae
gauf,
2005
PM 2
.5 E
mis
sion
Fac
tor (
g/kg
)
a a a
b
b
b
a denotes hard woods b denotes soft woods
Figure 2.3. PM2.5 emission factors from wood burning
Limited studies conducted to measured particle number emission factors from
biomass burning cause the lack of understanding of the impact assessments. There
has been only one study known to measure particle number emission factors of
biomass burning. The particle number emission factors of unspecified wood logs
were reported in the range of 1.43 to 39.5 ×1016 particles/kg depending on the
combustion system used (Wieser and Gaegauf, 2005).
2.4.7. Summary
Knowledge of particle emission factors from biomass burning is necessary in
estimating the impacts. The literature review of relevant studies reveals that most
measurements of emission factors were conducted in terms of mass emission factors.
Mass emission factors have been measured for different types of biomass, species of
biomass, and way of burning. The results show that those factors influence mass
43
emission factors. The available data for particle number emission factors is, however,
very limited.
2.5. Measurements of biomass burning particles
2.5.1. Fresh particles
Quantification of particle emissions from biomass burning has been conducted in
laboratories and the air, depending on the purpose of the measurements. Laboratory
and in situ measurements have been used to characterize particle properties from the
sources, or fresh smoke, and airborne measurements have been carried out to
characterize aged particles from biomass burning in the atmosphere. Laboratory
measurements of biomass burning particles were conducted by using stoves or fire
places (Todd, 1991; McDonald et al., 2000; Hedberg et al., 2002) and open fires
(Hays et al., 2002). Todd (1991) characterized the performance of several stoves
using wood of different hardness, and varying moisture content of wood. The study
showed significant relationships among these factors to the particle emissions.
Similar studies were conducted to investigate the relationship between the
performance of stoves and type of wood to particle emission factor from combustion
of the woods (McDonald et al., 2000; Hedberg et al., 2002). A mass emission rate
and emission profile of open burning of mixed types of woods containing a variety of
moisture contents, fuel arrays (configuration and geometry), compositions and
densities was performed in an enclosure (Hays et al., 2002). The result showed that
biomass species significantly contributes to the characteristics of particle emissions.
44
2.5.2. Aged particles
Particles from aged smokes have been measured in several regional areas to
characterize the properties of particles in the regions, including Africa (Anderson, B.
E. et al., 1996; Le Canut et al., 1996; Dubovik et al., 2002a; Eck et al., 2003;
Haywood et al., 2003), North America (Dubovik et al., 2002b; Eck et al., 2003);
South America (Andreae et al., 1988; Reid and Hobbs, 1998; Reid et al., 1998a;
Dubovik et al., 2002a; Eck et al., 2003), Europe (Fiebig et al., 2003), and the
Mediterranean (Formenti et al., 2002). Very few studies of the characteristics of
biomass particles in the atmosphere exist over the continent of Australia. Previous
measurements have been undertaken over the Eastern part of the continent (Gras,
1991). Few campaigns have focused on characterizing the biomass burning smoke in
both the Northern Territory of Australia and parts of Indonesia (Borneo) (Gras, 1999;
Tsutsumi, 1999). The reports showed particles measured in different regions have
specific characteristics. Particles measured from aged smoke in North America were
found to be larger than those in the tropics and sub tropics of Africa and South
America. The largest particles were measured to have a volume median diameter
(VMD) of 0.5 μm (Eck et al., 2003) and count median diameter (CMD) of 0.34 μm
(Formenti et al., 2002).
2.5.3. Particle measurement methods
Measurements of particles have been conducted in terms of mass concentrations and
size distributions. Measurements of mass concentrations are based on gravimetric
and impactor techniques, such as electrical low pressure impactor (ELPI), quartz
crystal microbalance (QCM)), tapered element oscillating microbalance (TEOM) and
45
DustTrak. Particle number concentration measurements were conducted using a
condensation particle counter (CPC). Particle size distributions were measured by
using complex techniques based on detecting the number and mobility of particles
with high accuracy, such as: Electrical low pressure impactor (ELPI), aerodynamic
particle sizer (APS) and scanning mobility particle sizer (SMPS) (Anderson, B. E. et
al., 1996; Le Canut et al., 1996; Formenti et al., 2002; Fiebig et al., 2003). Particle
size distribution was also measured based on diffusion principles, such as a diffusion
denuder and a diffusion battery (Fierz et al, 2002). Every measurement technique has
a specific characteristic in terms of resolution and accuracy. ELPI has high time
resolution of about 1 s with a large size spectrum (30 nm – 10 µm), however it has
limited size resolution; APS offers high time and size resolution in the size range of
larger than about 1 µm; and SMPS offers excellent size resolution in the range of 3–
1000 nm. The combining of a number of methods has been conducted to obtain
better accuracy and a wider range of measurement. SMPS and APS were used in
parallel for measurement of size distribution (Thomas and Morawska, 2002) to
obtain a wider measurement range compared to other techniques (Shen et al, 2002).
Measurements of particle size from biomass burning were conducted using available
techniques, these included impactor (IMP) (Echalar et al., 1998), optical particle
counter (OPC) (Reid et al., 1998a; Formenti et al., 2002; Fiebig et al., 2003) and
differential mobility particle sizer (DMPS) (Hobbs, P. V et al., 1996; Reid et al.,
1998a). SMPS was mostly used to measure size distribution of biomass burning
particles (Guyon et al., 2005b; Wieser and Gaegauf, 2005; Wurzler and Simmel,
46
2005). A combination of ELPI, APS, and SMPS was used to measure the size
distribution of particulate matter emitted from wood burning (Pagels et al., 2002).
The problem with measurement of particle concentrations from biomass burning is
that they are very high, whilst the available instrumentation has limitations in terms
of their measurement range. Another problem is that biomass burning particles
consist of compounds that grow to become bigger particles. In the case of
measurements of particles in real fields, there are other factors such as: field
conditions, fire intensity, and weather conditions contributing to the problem. A
technique addressing the problems is needed for measuring biomass burning
particles.
2.5.4. Summary
Measurements of particles from both fresh and aged smokes have been conducted in
laboratories and in the fields. Measurements of particles from biomass burning deal
with several problems due to characteristics of particles and high concentrations in
the field. However, a better method needs to be developed for investigating the effect
of various environmental conditions on the characteristics of biomass burning
particles.
2.6. Biomass Burning in Australia
In Australia, biomass burning occurs every year. Natural ecosystems, weather
conditions, landscapes, and even the country’s biological diversity sustain Australia’s
biomass burning. Climate is a factor that may increase the frequency, intensity and
47
size of biomass burning in Australia (Campbell, 2003). Variation of climate around
Australia affects the different frequencies of forest fires that mostly happen in the
zone dominated by dry eucalypt forests. The burnt areas of Australia are estimated at
about 6.5 % per year, except for forest fires in 1974 – 1975 that burned 15.2 % of the
continent (Luke and McArthur, 1978). According to the Australian fire report, about
115,000 to 230,000 fires per year have been counted by satellite remote sensing
during the fire seasons of 1998-1999 and 1999-2000, burning areas of 31 and 71
million ha respectively (Gill and Moore, 2005). The Western Australian Department
of Land Information reported that biomass burning across Australia from 1997 to
2003 affected areas of 26- 80.1 million ha. In this time, the greatest extent of biomass
burning was in the savannas of northern Australia. The Australian Bureau of
Statistics also reported that there were 5,999 forest fires consuming 21 million ha
across Australia, in which most area was in the Northern Territory of Australia with
15 million ha burnt, from July 2002 to February 2003 (ABS, 2004) .
The Northern Territory of Australia is a large savannah region suffering from fires
every year. The fires mostly occur during the dry season (Gill et al., 2000) with
mild intensity in the early dry season (EDS) and high intensity in the late dry season
(LDS) (Williams et al., 1998). Biomass burning in Northern Australian savannas for
the period 1997-1999 as estimated from interpretation of NOAA-AVHRR fire scar
mapping affected an area of 30 million ha (Russell-Smith et al., 2003a). During the
period 1997-2001, an average of 373,000 km2 of savannas in Northern Australia was
affected by fires. The worst fires occurred in 1999, consuming 4 million ha of land
(Russell-Smith et al., 2003b). Kakadu National Park in Northern Australia
48
experiences fires with 50 % of the area burned in the dry season every year. The data
from 1980 to 1995 showed that more than one million ha of the park area was
destroyed (Gill et al., 2000).
The state of Queensland is a part of Australia that also experiences fires every year.
It was recorded that the worst forest fires in 1991 consumed 37,000 ha (Hamwood,
1992). From July 2002 until June 2003, there were 2,618 fires in this states covering
one million hectares of land (ABS, 2004). In 2004, major fires occurred within the
South East corner of Queensland, including forest fires at San Fernando and Canugra
in July; at Wallaby Hill Mudgeeraba, Gold Coast, Minden, Gilston/Tallai Range and
Tamborine in August; and at Tamborine, Lowry/ Hinze Dam and Nerang in October
of that year.
2.6.1. Summary
Australia suffers from bushfires every year due to natural conditions such as the
natural ecosystem, weather conditions, landscape, and biological diversity.
Significant areas are consumed by biomass burning during Australian summer every
year. In particular, the Northern Territory and Queensland experience severe biomass
burning.
49
2.7. Biomass burning health impacts
Biomass burning has been known to have significant effects. The impacts on human
health have been presented by epidemiologic studies showing a relationship between
biomass burning and morbidity and mortality. A study in Southern Brazil reported
that the total concentration of particulate matter was significantly higher and the
patients requiring inhalation therapy also increased during the sugarcane burning
period (Arbex, 2000). Jacobs studied the relation between asthma patients in a
hospital and rice stubble burning for ten years. The report showed that the asthma
patients were 29 % higher than average on days during rice stubble burning (Jacobs,
1998). A five year study conducted in Japan found that the average number of
childhood asthma hospital visits were increased more than twice during the rice
burning months of September and October (Torigoe, 2000). Another study
presenting the association between sugarcane burning and hospital respiratory
patients reported that hospital respiratory patients increased about 50 % during
sugarcane burning season (Brill and Evely, 1994).
The impacts of biomass burning on human health can be seen from a toxilogical
prospective of the emissions. The World Health Organization WHO (1999) released
health guidelines for vegetation fire events. The guidelines identify the emissions
released from biomass burning as contributing to adverse effects on human health,
and divides them into a number of classes: particulate matter, polynuclear/polycyclic
aromatic hydrocarbons (PAH), Carbon monoxide (CO), aldehydes, organic acids,
semi-volatile and volatile organic compounds, nitrogen and sulphur-based
50
compounds, ozone and photochemical oxidants, inorganic fraction of particles, and
free radicals (Ward et al. 1989; WHO 1999).
2.7.1. Particulate matter
Particulate matter has been recognized as affecting human health, and is linked with
morbidity and mortality (Dockery et al., 1993; Schwartz, 1993; HEI, 2000). A lot of
studies have been done since 1990 to understand the relationship between particulate
matter, illness, hospitalisation, and premature death. A variety of the studies include
a number of topics, such as identifying cardiac responses to particles, the relationship
between particulate matter and asthma, elucidating possible biological mechanisms
for mortality, confirming the mortality effects around the world, and analysing the
effects in term of years, months, or even days.
0
5
10
15
20
25
0 20 40 60 80 100
PM10 Concentration (mg / m3)
Perc
enta
ge (
% )
120
Human deathHeart diseasePneumonia and COPD
Figure 2.4. The relationship between the PM10 concentration and morbidity and mortality (Data source: The Health Effects Institute, 2000; Burning Issues / Clean Air Inc, 2001).
51
The study conducted in 90 cities by the Health Effects Institute in 2000 showed a
significant relation between PM10 emission and morbidity and mortality. Figure 2.4
presents the relationships between the concentration of PM10 and the data of human
death, hospitalisation for heart disease, and hospitalisation for pneumonia and
chronic obstructive disease (COPD). It can be seen that for PM10 increases of 10
mg/m3, human death rises by 0.5 %, hospitalisation of heart disease increases by 1
%, and hospitalisation for pneumonia and chronic obstructive pulmonary disease
(COPD) goes up by 2 % (HEI, 2000). Samet and colleagues (2000) developed and
applied state of the art statistical techniques to analyse the relation between the
effects of pollutants including particulate matter on the extent of life shortening.
They found that a significant relationship existed between particulate matter and
mortality (Samet et al., 2000a; Samet et al., 2000b). Another study conducted across
20 cities in the United States showed the strong correlation between particulate
matter and hospitalisation among the elderly (Samet et al., 2000c). Levy et al
(2000), conducted a quantitative meta-analysis to estimate mortality from over
twenty daily time series studies. They found that mortality rates increased by 0.7 %
per 10 µg/m3 of PM10 concentration growth (Levy et al., 2000). Schwartz reported
that every 50 µg/m3 of particulate matter caused a 6 % increase in mortality and a
18.5 % increase in respiratory hospitalisation (Schwartz, 1993). Particulate matter
PM10 were also associated with post neonatal infant mortality from respiratory causes
(Woodruff et al., 1997; Bobak and Leon, 1999).
The relationship between PM2.5 emission and morbidity and mortality also has been
reported in previous studies. The study conducted by the American Cancer Society in
52
1995 reported on the association between fine particles PM2.5 and premature death
caused by cardio-pulmonary complaints. The report showed that the difference in
mortality was 17 % for a difference in PM2.5 between the cleanest and dirtiest cities
of 24.5 µg/m3 (Pope et al., 1995). Goldberg and colleagues (2000) used the details of
recorded health data from patients to analysis the correlation of particulate matter
and mortality in Montreal to link individual death to medical information up to five
years before death. The result showed that death related to cancer, chronic coronary
artery disease, and coronary artery disease was associated with the concentration of
PM2.5 (Goldberg et al., 2000). Other studies also showed the relationship between
particulate matter PM2.5 and morbidity and mortality caused by asthma (Vedal et al.,
1998; Norris et al., 1999; Tolbert et al., 2000; Yu et al., 2000), bronchitis (Etzel,
1999; McConnel et al., 1999; Peters, J.M et al., 1999) and heart disease (Peters, A et
al., 1999; Peters et al., 2000).
Epidemiologic studies have also provided evidence linking health effects and
exposures of ultrafine particles. The reports showed a strong association between
ultrafine particles and respiratory health in asthmatic adults (Peters et al., 1997; Von
Klot et al., 2000) and among children (Pekkanen et al., 1997). Positive correlations
of cardiovascular mortality with ultrafine particles were found in a epidemiological
study conducted by Wichmann and colleagues (Wichmann et al., 2000). It has been
shown that ultrafine particles have contributed to other epidemiological evidence of
adverse effects on the cardiovascular system (Oberdorster et al., 1995; Seaton et al.,
1999; Delfino et al., 2005). A study of the relationship between ultrafine particles
and mortality was conducted in Erfurt, Germany in 1995-98. The ultrafine
53
concentration and daily mortality were analysed using Poisson regression techniques
with generalized additive modelling (GAM). The result showed a positive
association between ultrafine particle concentration and mortality (Wichmann and
Peters, 2000).
Toxicological studies of ultrafine particles have been conducted based on their
capability of inducing inflammation per unit particulate matter mass due to their high
particle number, high lung deposition, and surface chemistry through reactive
oxygen species (ROSs) or other mechanisms to show their effects on human health.
It has been found that the deposition efficiency of ultrafine particles in human
subjects was found more than 60 % (Chalupa et al., 2004). Ultrafine particles have
been reported to be capable of inducing pulmonary inflammation, as well as entering
the cardiovascular system (Oberdorster, 2001; Nemmar et al., 2002; Oberdorster et
al., 2002; Nemmar et al., 2004). In terms of other toxicological basis, organic
compounds, such as PAHs, have been shown to induce a broad polyclonal expression
of cytokines and chemokines in respiratory epithelium. The previous studies showed
that ultrafine particles contain the largest fraction of polycyclic aromatic
hydrocarbons (PAHs) (Elguren-Fernandez et al., 2003; Li et al., 2003). In addition,
PAHs, metal, and other related compounds may lead to the production of cytotoxic
(ROSs) (Nel et al., 1998; Nel et al., 2001) that induce oxidant injury and
inflammatory response (Pritchard et al., 1996). There is evidence regarding the
importance of oxidant stress responses to cardiovascular effects (Dhalla et al., 2000).
Li and colleagues showed that ultrafine particles were most potent toward inducing
54
cellular heme oxygenase-1 (HO-1) expression and depleting intracellular glutathione
(Li et al., 2003).
2.7.2. Summary
Particulate matter in different size ranges has been recognized to have serious effects
on human health. Epidemiological and toxicological studies show the significant
relationship between particulate matter emission and morbidity and mortality.
2.7.3. Polycyclic aromatic hydrocarbons (PAHs)
Polynuclear/polycyclic aromatic hydrocarbons (PAHs) are known as carcinogenic
particles contributing to huge affects on human health especially in the development
of cancer (Arcos and Argus 1975; NRC 1983; Harvey 1991; WHO 1998, 1999).
Even mechanisms of PAH formation during the burning of organic matter is not still
fully understood, but the basic formation of PAH occurs when the radicals produced
by burning at high temperatures recombine to become PAH at lower temperature is
still a question (Simoneit 2002). PAHs include acenaphtene, acenaphthylene,
anthracene, benzo[a]pyrene, benz[a]anthracene, dibenz[a.h]anthracne, fluoranthene,
naphthalene, phenanthrene, and pyrene (WHO 1999; Morawska and Zang 2002).
Almost all PAHs are known as toxic. Benzo[a]pyrene is highly carcinogenic,
contributing to development of cancer in cells of humans. Benzo[a]anthrancene is
not only identified as a human carcinogen but also causes DNA damage and gene
mutation in mammalia cells.
55
2.7.4. Carbon monoxide
Carbon monoxide (CO) is an air pollutant produced by many incomplete burning
sources including biomass burning. Exposure of CO to humans causes
Carboxyhemoglobin that reduces the capacity of the red blood cells to absorb
oxygen, consequently people show signs of disorientation or fatigue (WHO 1999).
High concentration of carbon monoxide in the body can disturb the cardiovascular
system including the heart, lungs, blood vessels, and a gallon and a half of blood that
transports oxygen and removes carbon monoxide. Carbon monoxide can disturb the
rhythm of the heart in pumping the blood causing heart disease. The lungs consist of
millions of tiny sacks called alveoli that fill with air. The sacks adjust mechanism of
oxygen and carbon monoxide change. If the concentration of carbon monoxide is
high, the mechanism of oxygen and carbon monoxide will be disturbed. This may
cause lung disease such as asthma or bronchitis. The blood, a part of the
cardiovascular system, has a transportation function maintaining the body balance in
temperature, acidity, total fluid, and balance in the fluid. Blood acidity must be just
right for oxygen and mono oxide exchange to take place. Breathing high levels of
carbon monoxide creates acidity in the blood causing death of tissues and even
cancer (Rozenberg, 2001).
2.7. 5. Aldehydes
Aldehydes are chemical compounds also recognized as toxic gases, which are
extremely irritating to the mucous membranes of the human body. Formaldehyde,
one of the aldehydes produced by incomplete combustion, is transported rapidly in
the human body to form formic acid, which is removed very slowly. Exposure to
56
formaldehyde causes the ability of the cells of the lungs to engulf foreign bacteria to
decrease, which may accentuate infection of the respiratory system (WHO 1999).
Formaldehyde also can cause potent eye and skin irritation. Previous studies have
shown that formaldehyde is a potential human carcinogen, causing human cancer,
central nervous system effects including headaches, fatigue, and depression (Godish,
1989). Acetaldehyde, an aldehyde known to be toxic, generates olfactory epithelium,
liver lesions, and nasal cancer.
2.7.6. Organic acids
Acid compounds including organic acids are irritants. Several organic acids are
known as toxic and irritant. Acetic acid can irritate the skin or eyes. Formic acid is an
organic acid that is more of an irritant than acetic acid and dangerously caustic to the
skin. Exposure to organic acids via respiratory system can destroy the cardiovascular
system and accelerate cancer creation. Organic acids can also acidify the ground in
terms of acid rain (CA EPA, 2001).
2.7.7. Volatile organic compounds
Volatile organic compounds have serious effects on human health such as cancer and
other effects. Benzene is a dangerous volatile organic carbon causing mucous
membrane irritation, neurology symptoms, and death due to respiratory failure.
Exposure to benzene in high concentrations may result in bone marrow depression
and anaemia (Dorland, 1994). Benzene is also reported to cause reproductive
toxicity (CA EPA, 2001). Toluene gives the feeling of intoxication and causes huge
effects on human health, such as sleepiness, dizziness, headache, muscular weakness,
57
confusion, impaired co-ordination, generation of respiratory tract skin, erosion of the
nose and changes in the liver and kidneys. High concentration exposure causes
damage to the brain stem.
2.7.8. Dioxin
Dioxin is a toxin which is a very potent carcinogen and endocrine disrupter. Dioxin
released by biomass burning or other pollutant sources into the atmosphere as smoke
particles, fall to the earth in rain, lodge in the soil and water, and coat water and land
plants. Plants contaminated by dioxin are eaten by animals. Humans eat food coming
from the plants and the animals and drink water that is also contaminated by dioxin.
The recycled processes are repeated over and over again in the environment for
years. Dioxin lasts in the human body for seven years. Chlorinated dioxin causes
serious problems with the immune and endocrine systems in humans, foetal
abnormalities or death (CA EPA, 2001).
2.7.9. Elementary elements
This section briefly reviews several elements mostly emitted by biomass burning,
such as lead (Pb), manganese (Mn) and aluminium (Al). Lead (Pb) is a toxic
compound poisoning humans, with symptoms of headache, insomnia, dizziness,
hypertension, albuminuria, amenia, loss of appetite, and constipation (CA EPA,
2001). Lead also causes brain and other nervous system damage and is identified as
decreasing the intelligent quality of children. Manganese (Mn) is a toxin acting on
the central nervous system that causes impairment of neurobehavioral functions, such
as slowed visual reaction time, erratic fine hand, forearm movement, and finger
58
tremor. It also causes respiratory problems, including irritation of respiratory systems
and acute bronchitis (Holmes et al., 1989; Dorland, 1994). Exposure to aluminium
(Al) may cause pulmonary fibrosis and neurological symptoms that may be fatal if
the amount of aluminium in the bloodstream is excessive (CA EPA, 2001). Other
detected inorganic compounds emitted by biomass burning are also toxic: nickel
(Ni), Zinc (Zn), Copper (Co), etc.
2.7.10. Summary
Biomass burning produces particle and gaseous emissions which may pose
considerable environmental and health risks. More than 300 chemical compounds
released from biomass burning may have serious effects on human health.
2.8. Dispersion model
2.8.1. Theoretical background
Knowledge of transport mechanisms of particle emissions from a source to a receiver
is important in order to get a better understanding of the impact assessments on
environment and human health. However transport mechanisms of particle emissions
in the atmosphere is very complex due to natural conditions and physical, chemical,
and thermodynamic processes during their transport. Transport mechanics of
particles has been studied based on field analysis and complex theoretical
approaches. A basic theory used to explain transport mechanism is called diffusion,
which is a process that occurs when particles diffuse and suspend to a host substance
having a similar size (Csanady, 1980). The diffusion process of particles, which is
represented as mass concentration (g/cm3), and number concentration (particles/cm3)
59
may differ at different points in space. Diffusion of particles occurs if the
concentration is higher on one side of a boundary than on the other. Diffusion rate or
diffusion flux is generally expressed by Fick’s law:
F = - D∇C (2.1)
where D is diffusivity constant and C is concentration.
Based on Fick’s law, the classical diffusion equation is derived to obtain a general
dispersion equation
[ CK Cu ]tC
∇∇=∇+∂∂ (2.2)
where tC
dd is the change of concentration in time and u is the speed vector,∇ is the
gradient operator and K is the Eddy diffusivity constant.
Equation (2.2), known as the Navier-Stoke equation, describes concentration as a
spatial and temporal function that includes vertical and horizontal transport
(advection) and vertical and horizontal diffusion. Based on the equation, the
concentration of particles in the atmosphere can be estimated. Particles released from
a source into the atmosphere are usually considered as a plume. A transport model of
the emission concentration in a plume is known as a plume dispersion model or
dispersion model. A variation of the dispersion model depends on an analysis
approach of a plume. A dispersion model is derived based on a number of factors:
homogeneity of the atmospheric boundary layer (ABL), homogeneous (Kim and
Larson, 2001; Moissette et al., 2001) and non-homogeneous (Lines et al., 1997;
Heinz, 1998; Heinz and Dop, 1999; Ermak and Nasstrom, 2000; Carvalho et al.,
2002; Ferrero et al., 2003; Franzese, 2003; Iliopoulos et al., 2003); stability
condition of ABL (Kim and Larson, 2001; Mangia et al., 2002); phase of
60
pollutants: particles (Davidson et al., 1995; Gouesbet and Berlemont, 1998; Du,
2001; Kim and Larson, 2001; Malcolm and Manning, 2001; Mangia et al., 2002;
Ditlevsen, 2003; Franzese, 2003; Iliopoulos et al., 2003) or gas (Lines et al., 1997;
Seland and Iversen, 1999; 2001; Flemming et al., 2001; Oettl et al., 2001; van Baten
and Krishna, 2001; Carvalho et al., 2002; Chahed et al., 2003; Tsuang, 2003); and
physical and chemical processes (Owczarz and Zlatev, 2002; Tsuang, 2003).
2.8.2. Classification of dispersion model
The dispersion model may be classified into three categories based on the theoretical
approach to the plume: Eulerian model, Lagrangian model, and Statistical
Particle model. The Eulerian model treats a plume as containing coupled boxes or
grid cells with the assumption that particles are uniformly mixed in a box or cell
(Sportisse, 2001; Moschandreas et al., 2002; Chahed et al., 2003). The distribution of
particles is described by changing the concentrations at discrete points in a cell.
Concentrations and flux of particles passing in a cell are usually presented as a
function of space (Christensen, 1997; Seland and Iversen, 1999; Zhou and
Leschziner, 1999; Ulke, 2000; Becker et al., 2001; van Baten and Krishna, 2001;
Mangia et al., 2002). The Lagrangian model approaches a plume as a dispersion
story of individual particles transporting along a trajectory. The change of
concentration is described as a probability of particles that diffuse at a certain time.
The Lagrangian model can be classified into steady state Lagrangian model and
Lagrangian segmented or PUFF model. The Lagrangian model executes a story of
particle dispersion in a certain time at a steady state condition. The concentration
and flux are calculated by integrating a dispersion equation over a certain range of
61
time (Ermak and Nasstrom, 2000; Franzese, 2003). For PUFF Lagrangian models,
the story of the emission dispersion is approached in small periods of time. In other
words, a plume is segmented into small pieces (Souto et al., 2001; Carvalho et al.,
2002). The concentration and flux are calculated by summing the concentration and
flux of the segmented plume. The Statistical particle model considers a plume as a
series of particles in which individual particles move randomly along a trajectory or a
grid cell (Zhou and Leschziner, 1999; Du, 2001). The concentration and flux are
calculated by summing the probability of emission concentrations along a trajectory
or in a cell.
Lagrangian model
The Lagrangian model assumes that particles move through space and time. The
model approaches a plume as individual particles travelling along a trajectory. The
concentration is described as the probability to determine the particles that diffuse in
space at a certain time. The concentration of particles is commonly calculated using a
probability density function (PDF). If the distribution concentration follows a normal
or Gaussian shape, the model is called a Gaussian model (Jennings and Kuhlman,
1997; Oettl et al., 2001; Raza et al., 2001; Venkatesan et al., 2002; Tsuang, 2003).
The concentration can be also calculated by treating a plume as a segmented puff at
which each puff is separated independently. The final concentration at a certain time
is a superposition of the concentrations of all puffs. This model is called the PUFF
model (Lines et al., 1997; Souto et al., 2001; Jung et al., 2003). If the distribution of
concentrations along the trajectory is Gaussian, the model is a Gaussian PUFF
model.
62
A basic statistical concept may be applied to describe particle movement. The
concept, called probability distribution, is used to describe particles that randomly
move in space and time. The random movements are presented by a probability
density function (PDF) that describes the probability to determine particles that
displace from one place to another. For example, let a particle be at position x0 at
time t=0, and at x at a later time t. Its displacement may be represented by the
probability density function P(x – x0, t) that describes the probability of finding a
particle at position x and time t in a volume of dx. The chosen form of the PDF
represents realistically the actual behaviour of the particles in the space.
Probability of a particle at a certain position is commonly determined using the
concept of concentration (C) defining a number of particles in a volume. An
ensemble mean concentration of particles at position (x, y, z) at time t may be
defined as,
C (x, y, z, t | x0 , y0, zo ) = Q P (x – x0 , y – y0 , z – z0 , t ) (2.3)
In terms of integration form, the ensemble mean concentration of particles can be
also presented as follows:
dxdydzdt t),z-z,y-y,x-x( P Q)z,y,x t z, y, (x, C 0ooooo ∫ ∫∞−
∞
∞−
=t
(2.4)
Equation (2.4) shows a correlation of particle existence at a future time related to the
existence of a particle at a previous time. This is known as a stochastic process. The
stochastic process may be characterized by a correlation function (R). For example, a
particle moves constantly along the x axis, the velocity at time t is defined as u (t),
63
and u (t + τ ) is the velocity at given delay time τ. The correlation function of the
velocities is presented as (Csanady, 1980),
2(t)u
) (t u (t)u )(R ττ += (2.5)
Where R (τ) has a value between –1 and 1. The velocity correlation function is an
important function on the stochastic process, associated with the particle position
before and after movement. Several studies applied a velocity correlation function in
order to determine a possibility of particle displacement in a Lagrangian stochastic
model (Heinz and Dop, 1999; Degrazia et al., 2000; Ermak and Nasstrom, 2000; Kim
and Larson, 2001; Carvalho et al., 2002; Ditlevsen, 2003; Ferrero et al., 2003;
Franzese, 2003; Iliopoulos et al., 2003).
The way to predict the probability of particle position in a space at a certain time
varies for Lagrangian models depending on analytical approaches. A number of basic
theories have been developed in order to estimate displacement of particles at a
certain time: Brownian motion (Taylor, 1921; Csanady, 1980; Gouesbet and
Berlemont, 1999; Becker et al., 2001; 2002) , random work (Chandrasekhar, 1943;
Cramer, 1946; Bartlett, 1956; Feller, 1957; Ditlevsen, 2003) and Monte Carlo
(Ermak and Nasstrom, 2000; Kim and Larson, 2001; Souto et al., 2001). All have
been used to develop modern dispersion models.
Brownian motion uses the assumption that particles are much larger than the
molecular structure of the surrounding fluid, and particles are small enough to be
influenced by molecular collision (Csanady, 1980). In order to obtain a detailed story
64
of distribution of particles, the total displacement is divided into a number of
independent steps in which each step is taken at random and independent from any
previous step. This method is called a random walk. The distribution probability of
random walks needs more complex calculation. The basic theoretical approach of a
random walk, introduced by Chandrasekhar (Chandrasekhar, 1943) and found in a
number of literature sources (Cramer, 1946; Bartlett, 1956; Feller, 1957) has been
used in dispersion models (Weil, 1990; Thomson and Montgomery, 1994).
Ditlevsen (2003) used a random walk model to solve the diffusion equation to
determine an accurate approximation of the probability density function of particle
position (Ditlevsen, 2003).
Particle displacement based on a random walk model is formulated by supposing a
displacement in a certain direction that is divided into a number of steps. For
example, the displacement in the x direction is divided into m steps at which each
time step jumps a fixed distance r forwards and backwards along the x axis. The time
needed to displace from one position to another is also divided into N time steps. The
probability to find a particle released at position x = 0 at an arbitrary position x is
presented in a probability density function as follows (Csanady, 1980; Sherwin,
1999),
⎥⎦
⎤⎢⎣
⎡−=
N 2mexp
N 2 N) (m, P
2
π (2.6)
Where N is the number of time steps or jumps that have occurred and m is an integer
such that x = r × m. The concentration of a particle at position x and time t = N Δt
where Δt is the time step between jumps is given as,
65
⎥⎦
⎤⎢⎣
⎡−=
N2xexp
N2
2rQ N)(x, C
2
π (2.7)
The Monte Carlo method is another statistical approach that has been used to
estimate particle displacement by tracking particles along a trajectory, in which each
trajectory is independent from the others (Zannetti, 1984; Cogan, 1985; Zannetti,
1990; Boybeyi and Raman, 1995; Boybeyi et al., 1995; Ermak and Nasstrom, 2000;
Kim and Larson, 2001; Souto et al., 2001; Venkatesan et al., 2002). Basically, Monte
Carlo calculates an individual particle trajectory at which the particle displacement is
governed by the local mean wind speed (transportation) and the turbulent velocity
fluctuation (diffusion). The particle positions are presented as follows,
t (t) w t (t) w z(t) t) z(t
t (t) v y(t) t) y(t t (t)u t (t)u x(t) t) x(t
Δ′+Δ+=Δ+Δ′+=Δ+
Δ′+Δ+=Δ+ (2.8)
Where w and u represent the horizontal mean and vertical velocity respectively,
while are the turbulent velocity fluctuations, t is the time and Δt is the
time increment. Estimation of the turbulent velocity fluctuations may become a
critical issue because the components are semi random. The components were
randomized by manipulating random numbers obtained from a first order
autocorrelation process or Markov process (Smith, 1968; Hanna, 1979; Souto et al.,
2001). They used an autocorrelation function, R, which was related to the Lagrangian
time scale by R(Δt) = exp (- Δt/ T
w and ,v ,u ′′′
L), where the Lagrangian time scale TL was
estimated for each component.
66
Another method to estimate the turbulent velocity fluctuation is by using Bachelor
theory combined with a relative diffusion correlation function (Tombrou et al.,
1998). Zanneti (1984; 1990) presented the semi random turbulent velocity fluctuation
as summing auto correlation velocity fluctuation components with purely turbulent
velocity fluctuations,
t) (t w t) (t u (t)w t) (t w
t) (t v (t)v t) (t vt) (t u (t)u t) (t u
43
2
1
Δ+′′+Δ+′+′=Δ+′Δ+′′+′=Δ+′Δ+′′+′=Δ+′
φφφφ
(2.9)
Where w and ,v ,u ′′′′′′ are purely random, independent, and uncorrelated turbulent
velocity fluctuations.
Homogeneity of media associated with a stability of media determines selection of a
statistical method in Lagrangian models. Behaviour of the particles in homogenous
media is simpler than that of inhomogeneous media. The probability of finding
particles in the space at a certain time is obtained by determining the probability of
particle speed that will be constant at any time. On the other hand in inhomogeneous
media, to determine particle dispersion in a certain time needs more complicated
analysis due to a variation of the speed of particle dispersion in space and time.
Consequently a complicated statistical method may be used, and an extra careful
choice in preferring an available statistical method is needed.
In general, a Lagrangian model is simple in describing particle dispersion by
predicting the transport of particles along a trajectory. The model is good for
recording a track of particles coming from a major source. The model is powerful for
estimating a particle growth or secondary particle (Malcolm and Manning, 2001) and
67
first-order chemical reaction (Calori and Carmichael, 1999; Koe et al., 2001).
Lagrangian models work well both for homogeneous and stationary conditions over
flat terrain (Jennings and Kuhlman, 1997; Oettl et al., 2001; Raza et al., 2001;
Venkatesan et al., 2002; Tsuang, 2003) and in inhomogeneous and unstable media
conditions for complex terrain (Du, 2001; Hurley et al., 2003; Jung et al., 2003). The
numerical computation of Lagrangian models is relatively simple and far from the
computational complexities associated with the simultaneous solutions of many
different equations (Stohl, 1998). Lagrangian models do not require numerical
differential techniques to calculate the mean concentrations (Nguyen et al., 1997).
The model usually requires less computational resources, needing only a personal
computer and less computational expertise.
Lagrangian models are appropriate to be applied for a point source of pollutant. The
model needs more complicated approaches for other types of pollutant sources such
as line, area, and volume sources. Lagrangian models are unsuitable for describing
high-order or non-linear chemical reactions, which commonly occur in the
atmosphere.
Eulerian model
The basic concept of the Eulerian model is that a plume is approached as a fixed grid
cell at which the concentration is defined. The concentration is described at a given
position and time, with the coordinates fixed in space and time as follows,
C utC
∇+∂∂ = - [ CK∇ ]∇ + S (2.10)
68
where C is the concentration and u is the velocity vector. The first term on the right-
hand side represents turbulent transport of the trace pollutants. K is the eddy
diffusivity and S is the source term. A number of theories, including K and gradient
transport theory, were used to determine the concentration of particles at a different
location of the grid and time for a certain condition (Pasquill and Smith, 1983;
Gouesbet and Berlemont, 1999; Ulke, 2000; Sportisse, 2001; Mangia et al., 2002).
The concentration of particles for a point of source may presented either as a spatial
derivative (Christensen, 1997; Sportisse, 2001)
⎥⎦⎤
⎢⎣⎡
∂∂
∂∂
=∂∂
zCzK
ztC )( (2.11)
or a temporal derivative (Pasquill and Smith, 1983; Ulke, 2000; Sportisse, 2001;
Mangia et al., 2002):
⎥⎦⎤
⎢⎣⎡
∂∂
∂∂
=∂∂
zCK(z)
zxCu(x) (2.12)
where C is the crosswind-integrated concentration, u(x) is the horizontal wind speed,
K(z) is the eddy diffusivity describing atmospheric turbulence, z and x are the
vertical coordinate and the along wind direction respectively. zCK(z)∂∂ represents
the vertical turbulence flux, showing a turbulence effect that depends on a stability of
the atmospheric boundary layer (ABL). Stability of ABL is reflected by homogeneity
or non-homogeneity of the atmosphere. For stable boundary layer shown by small
eddy turbulent diffusivity, the shear driven ABL is influenced by only mechanical
turbulence force; whilst in an unstable boundary layer, mechanical and convective
forces generate a turbulence (Ulke, 2000; Mangia et al., 2002). Consequently the
eddy turbulence diffusivity is large for unstable boundary layers.
69
A scale of the grid cell in a Eulerian model may be taken on a large scale to simplify
the model. By summing a scale is large and a condition in the grid cell is
homogenous or a perfectly stirred reactor (PSR), the Eulerian model becomes a
Eulerian Box model or Box model. A box model is a simple dispersion model to
describe the transport mechanism of particles in the atmosphere. To simplify
calculation of the concentration, the Eulerian model may be derived with the
assumption that the vertical scale of the box is equal to a mixing height, and the
condition in the box is perfectly homogenous. However, the real condition of the
ABL is far from the homogeneous condition. A dispersion study of several species
in the atmosphere using the model experienced a high error (Sportisse, 2001).
Eulerian model has advantages in predicting interactions between liquid and gas
phases (Zhou and Leschziner, 1999; van Baten and Krishna, 2001; Chahed et al.,
2003). The interaction of the phases is predicted by taking into account turbulenct
force for a mass balance condition. The simulation of the interaction was
demonstrated by using computation fluid dynamics (CFD) (Zhou and Leschziner,
1999; van Baten and Krishna, 2001). The shear effect of particle flow was handled
well in the model. This was useful to predict the stability of the atmospheric
boundary layer (Ulke, 2000; Mangia et al., 2002). The Eulerian model
accommodates non-linear chemical reactions, which are necessary to estimate the
concentration of emissions, including ozone concentration (Flemming et al., 2001;
Meith et al., 2003). The model offered the best approach for the future atmospheric
70
model due to the inclusion of a high-order chemical reaction (Peters et al., 1995;
Nguyen et al., 1997; Stohl, 1998).
Disadvantages of Eulerian model include a difficulty in determining a source,
because the model assumes that particles are continually distributed in the grid cell.
The model has difficulty identifying the impacts of every single pollutant on the
receptors, because there are no solutions for the total concentration that describe the
contribution of the pollutant to the receiptors. The Eulerian model is complicated in
numerical solution or in computation as it includes many differential equations. In
particular, the model is used to model mesoscale photochemical processes at which a
single grid representing the concentration of components which can not to be
reduced in a numerical computation (Nguyen et al., 1997; Stohl, 1998). The model
needs not only high expertise in computing but also a very high-powered computer
(Nguyen et al., 1997; Stohl, 1998; Moschandreas et al., 2002). A numerical diffusion
using a Eulerian model is just a conceptual approach rather than physical diffusive
processes (Chock, 1991; Yamartino et al., 1992; Odman and Russell, 1993).
Statistical particle model
A Statistical particle model treats a plume as a series of particles or sometimes
thousands of particles in which each individual particle is transported separately.
Concentration of each particle is estimated by determining a probability of the
displacement as a function of time. Probability of particle displacement may be
determined by tracking of the particle position within a fixed cell as in the Eulerian
approach (Zhou and Leschziner, 1999) or by tracking along particle trajectories as in
71
the Lagrangian approach (Du, 2001). Statistical particle models are very complex
and expensive in terms of computing, needing at least 105 particles to represent a
pollutant source. The model is very useful to present the concentration of pollutant
close to the source.
2.8.3. Dispersion model for biomass burning
Transport mechanisms of biomass burning emissions are very complicated due not
only to some factors influencing the emission production, but also natural conditions
and atmospheric processes occurring during the transport. Type of emission
(particulate matter or gas) determines the approach of the dispersion model. To
model particulate matter dispersion differs from gases in the models, even though a
few of the dispersion models consider gases to be particles. The percentage of the
emissions (particulate matter and gases) is another factor that should be included in a
biomass dispersion model. Source factors such as fuel type, fuel moisture, heat rate,
emission factor, and emission rate should be taken into account in a dispersion model
for biomass burning. The burning phases of flaming and smouldering are unique
factors contributing to the complexity of the dispersion model. Meteorology
conditions (temperature, humidity, wind speed, and wind direction) and mixing
height should be included in a dispersion model for biomass burning. Physical and
chemical processes during the transport of emissions in the atmosphere and the
complexity of terrain are other factors adding to the complication of dispersion
model for biomass burning.
72
Several available dispersion models are used to estimate the distribution of emission
concentrations from biomass burning, such as SASEM, VALBOX, VSMOKE-GS,
NNFSPUFF, TSARS+, CALPUFF. The models have been adapted for other
purposes, for example, industrial modelling (Breyfogle and Ferguson, 2003), volcano
eruption modelling (Trentmann et al., 2002) and oil fire modelling (McGrattan,
2003). Breyfogle and Ferguson (2003) studied several dispersion models for
biomass burning to examine the components and efficiency for implementation.
They concluded that the available dispersion models were basically applicable but
they could not be valid and suitable for biomass burning. Treatmann et al (2002)
simulated the transport mechanism of a smoke plume based on the information of the
emissions and atmospheric conditions. They used the active tracer high-resolution
atmospheric model (ATHAM) originally designed to simulate explosive volcanic
eruptions (Oberhuber et al., 1998), and compared the result with airborne remote
sensing and in situ measurements. They reported that ATHAM was a reliable model
to estimate dispersion of biomass burning emissions. They concluded that ATHAM
was a potential dispersion model of biomass burning emissions if it would be
combined with a microphysical and chemical model (Trentmann et al., 2002).
The basic dispersion models for biomass burning are the Lagrangian Gaussian
model, Lagrangian PUFF model, Eulerian model, and Eulerian Box model. The
dispersion model SASEM (Sestak and Riebau, 1988) and VSMOKE-GIS are
Lagrangian Gaussian models that calculate plume trajectories which vary in time and
space with the assumption that the concentration in a plume at crosswind is Gaussian
(bell-shape). Dispersion models such as NNFSPUFF, CALPUFF (Scire et al., 1995)
and TSARS+ are Lagrangian Puff models treating a plume that moves along a
73
trajectory as series of puffs (Hummel and Rafsnider, 1995). The dispersion model
ATHAM (Oberhuber et al., 1998) is a Eulerian model dividing a plume into grid
cells. The model VALBOX (Sestak et al., 1989) is a Eulerian Box model.
Most of the dispersion models for biomass burning were used for total suspended
particles (TSP) and PM10. The dispersion models NNFSPUFF, TSARS+ and
ATHAM were used for PM2.5 (Hummel and Rafsnider, 1995; Scire et al., 1995;
Trentmann et al., 2002). The dispersion model ATHAM, which was theoretically
designed to estimate the distribution concentrations of particles, water vapour, and
gas, was used to model the dispersion of PM2.5 from a smoke plume (Trentmann et
al., 2002). The dispersion models NNFSPUFF, TSARS+, and VALBOX were
applied in modelling concentrations of CO. The dispersion model TSARS+ was also
used to calculate concentrations of CO2 and CH4.
Topology of terrain is a factor that has been taken into account in a dispersion model
for biomass burning. Complexity of terrain is related to forced surface and mixing
height. Complexity of terrain also determines the surface wind that influences
turbulence effects and has an impact on emission rate during the burning process in
its flaming and smouldering phases. The dispersion models SASEM and VSMOKE-
GIS were designed for relatively flat terrains. The dispersion model VALBOX was
designed for a simple valley terrain. The other models NNFSPUFF, CALPUFF,
TSARS+, and ATHAM were aimed to calculate the distribution concentration in
complex terrains.
74
Table 2.2. Characteristics of the dispersion models for biomass burning. Model Dispersion
Model Burn Sources
Emissions Range Terrain Source Strength
SASEM Lagrangian Gaussian
Single PM10, TSP 100 km Flat EPM or internal SASEM
VSMOKE-GIS
Lagrangian Gaussian
Single PM10, TSP 100 km Flat Internal VSMOKE-GIS
NNFSPUFF Puff Multi PM2.5, PM10, TSP, CO, CO2, CH4
488 kma Complex EPM
CALPUFF Puff Multi PM10, TSP a Complex EPM TSARS+ Puff Multi PM2.5, PM10,
TSP, CO, CO2, CH4
300 kma Complex EPM or internal SASEM
ATHAM Eulerian Single PM2.5 100 km Complex EPM VALBOX Eulerian
Box Multi PM10, TSP, CO 900 mm
feet Simple Valley
EPM or internal SASEM
a Limited only by available domain
2.8.4. A model for biomass burning study
Modelling transport mechanisms of biomass burning emissions is a difficult task due
to complex factors playing important roles in transporting the emissions from sources
to receiptors. The factors can be categorized as source factors, meteorology factors,
and dispersion model approach. Source factors are associated with emission
production that consists of amount of emissions, emission factors, and energy rates.
Meteorology factors, such as wind speed, wind direction, humidity, and temperature,
contribute important aspects in transporting biomass burning emissions in the
atmosphere. Dispersion model approach used to estimate the concentration
distribution of the emissions in the atmosphere is the most important step in
modelling of emission transport, especially related to the impact assessments.
Source Strength Models
A source factor is a parameter associated with emission production of biomass
burning. Type of emissions, emission factor, emission rate, and heat rate should be
included in modelling source factors. It is known that biomass burning emits
75
particulate matter and gasses (Muraleedharan et al., 2000; Oros and Simoneit, 2001;
Anderson, 2002; Dennis et al., 2002; Hays et al., 2002; Hedberg et al., 2002; Khalil
and Rasmussen, 2002; Pagels et al., 2002; Reddy and Venkataraman, 2002b). The
emission factor from biomass burning has been recognized as depending on type of
vegetation (Oros and Simoneit, 2001; Hays et al., 2002; Hedberg et al., 2002;
Edwards et al., 2003; Gullett and Touati, 2003; Khalil and Rasmussen, 2003), fuel
moisture (Core, 1984; Oros and Simoneit, 2001; Hays et al., 2002; Hedberg et al.,
2002; Edwards et al., 2003; Gullett and Touati, 2003; Khalil and Rasmussen, 2003) ,
and speed of burning (Hueglin et al., 1997; Zou et al., 2003).
Heat rates, including heat intensity (heat released per second) and severity (heat
release per unit area), are other source factors that should be estimated in the
transport mechanism of emissions. These factors are related to convective energy
used to transport the emissions vertically before they disperse horizontally. More
heat released means the higher the emissions are raised in the air. The height of the
emissions causes the mixing height of the atmosphere boundary layer to become
larger, which affects the dispersion process of the emissions. Amount of burnt fuels,
burning area and weather condition are factors affecting heat rate.
Several source strength models have been developed to calculate biomass burning
emissions, such as the emission production model (EPM) (Sanberg and Peterson,
1984), and internal algorithms included in the dispersion models SASEM,
VSMOKE-GIS (Sestak and Riebau, 1988), CONSUME (Otmmar et al., 1993), and
BurnUp (Albini et al., 1995; Albini and Reinhardt, 1995; 1997). Emission production
76
models (EPM) were designed to simulate heat emission rate, gas emission and
particulate matter in wild land fires corresponding to the rate of fuel consumption
(Sanberg and Peterson, 1984). An EPM estimates emissions of biomass burning
under a variety of burning conditions, types of fuel, and ignition patterns; however
the model does not take into account emission rate for the smouldering process.
Some emissions are produced in significant amount during the smouldering phase
(Hays et al., 2002). The model CONSUME is a source strength model that estimates
emissions and total emission in flaming and smouldering stages by assuming the
burn areas have a homogeneous distribution of fuel elements. EPM coupled with
CONSUME was linked to several dispersion models (Breyfogle and Ferguson,
2003). The model BurnUp was also used to estimate emission rates of flaming and
smouldering phases for biomass burning. The model BurnUp coupled with a fire
spread model FARSITE (Finney, 1998) was used to study wildfire in the Northwest
United States (Ferguson et al., 2001).
Meteorological Models
Meteorological data on wind speed, wind direction and temperature are called core
meteorological parameters, and are fundamental inputs for a dispersion model. A
steady state Gaussian dispersion model needs hourly meteorological data at the
surface and an upper station to estimate the mixing height. More sophisticated
dispersion models, such as PUFF models and Eulerian models, require not only core
parameters but also addition meteorological parameter including atmospheric
pressure, humidity, precipitation, turbulence parameters, solar radiation, and land-sea
temperature data (Moschandreas et al., 2002).
77
Availability of meteorological data determines the accuracy of a dispersion model.
The meteorological data should be available for the place where the measurement of
the concentrations takes place. Sufficiency of meteorological data also influences the
accuracy of dispersion model. Sufficiency of meteorological data at the surface and
upper air over modelling domains is shown by representative meteorological stations,
adequate instruments and skilled personnel. The accuracy of a dispersion model
becomes less when the observed meteorological data are taken from distant stations
that may not represent the condition of the measurement area. It may even result in
error, when the observed meteorological data are unavailable.
Where observed meteorological data are incomplete or unavailable; there is an
alternative method to establish the meteorological conditions that is called the
numerical weather prediction (NWP) model. The model outputs wind speed, wind
direction, temperature, humidity and other parameters to cover the domain. The
model can be coupled with local geophysical features to produce localized
meteorological fields. The mesoscale Model Version 5 (MM5) is a NWP model that
was applied for a study in western Canada, whilst the meteorological data were taken
in British Columbia, and Alberta (Grell et al., 1994).
Dispersion Model Approach.
To select a model from the available dispersion models for biomass burning study is
not easy. Every dispersion model is originally used for a different case and has
78
strengths and weaknesses. In order to determine an appropriate model for this study,
a number of criteria are set up:
• The model should be able to realize the objectives of the study
• The model should calculate the factors contributing to emission dispersion
• The model should include physical processes
• The model should be consistent
• The model should have good resolution
• The model should be suitable for the topology of the study area
Objective of Study
A dispersion model should fulfil the objectives of the study, that is to get a better
understanding of the characteristics of biomass burning particle emissions and their
impact assessments. The dispersion model should have the capability to estimate the
concentration distribution of particles from sources to receiptors. The dispersion
model should be presented as a function of space and time. Due the study’s focus on
ultrafine particles, the model should have the capacity to describe the transport from
sources to receiptors. The model should include some possibilities of physical
processes happening during the transport. The dispersion models that are suitable for
this purpose with modification are NFSPUFF, TSAR+, ATHAM, SASEM,
VSMOKE-GIS, CALPUFF, and VALBOX.
Physical Processes
Particles emitted by biomass burning overcome physical processes during their
transport in the atmosphere. These physical processes, including coagulation,
79
deposition, condensation and radioactive decay, occur during the transport from a
source to a certain height (particle rise) and during their dispersion in the
atmosphere. To apply the available dispersion model a number of modifications were
required, based on the characteristics of ultrafine particles, assumptions made,
empirical coefficients, theoretical approach, and computational analysis.
Complete model
Because biomass burning is a complex process, modelling transport mechanisms of
particles is very complicated. Several factors that contribute to transporting biomass
burning particles from a source to a mixing height called emission rise factors and
dispersing particles called emission dispersion factors have to be taken into
account. Emission rise factors are associated with factors corresponding to emission
production that consists of emission components, emission factors, emission rates,
and heat rates (heat intensity and heat severity) at which those factors cause the
emissions to rise to a certain high called a mixing high. Some factors playing a role
in emission production, such as type or species of biomass, heat components, ignition
pattern, fuel moisture, fuel loading, and local weather, should be calculated
accurately in the model. The model presenting the calculation of emission production
as a function of the influenced parameters is named the strength source model.
Horizontal transport (dispersion) of biomass burning particles should be presented in
the model. The model should involve calculation of some parameters affecting the
particle dispersion. Terrain complexity and weather condition are factors influencing
the dispersion that should be included in the model. Topology of terrain generally
80
affects wind speed and wind direction having an impact on the emission dispersion.
Complex terrain creates wind circulation that causes aged emission to return to mix
with fresh emissions.
Weather conditions shown by meteorological data, such as wind speed, wind
direction, temperature, humidity, atmospheric pressure, precipitation, turbulence
parametesr, solar radiation, and land sea temperature; play important roles in particle
transport in the atmosphere. Core meteorological data of wind speed, wind direction,
and temperature should be available and sufficient. A numerical weather prediction
(NWP) model may be used to estimate meteorological data. The model may be
coupled with local geophysical features to produce local meteorological fields.
A dispersion model selected based on this criteria should contain a strength source
model and a meteorological model as integrated parts. The available dispersion
models for biomass burning have a strength source model. The dispersion models
SASEM, NFSPUFF, CALPUFF, TSARS+, ATHAM, and VALBOX use an emission
production model (EPM) as their strength source model. The dispersion models
SASEM, VSMOKE-GIS, TSARS+, and VALBOX have their internal strength
source model.
Consistency of Model
Consistency of a dispersion model, which is shown in deriving a formula and
calculating some factors describing physical processes that happen during the
transport, should be considered in choosing an available dispersion model or
81
developing a dispersion model. Consistency in calculating a turbulence field of the
air is also used as a criterion in choosing a dispersion model. A turbulence field
corresponding to an emission velocity is a crucial factor that plays an important role
in transporting and dispersing particles. The dispersion of particles from the sources
to the receiptors should be traced consistently in the dispersion model. Theoretical
and analytical approaches establishing the frame should be reliable in formulating the
dispersion model. The dispersion models that fulfil this criterion are SASEM and
VSMOKE-GIS (Gaussian Lagrangian), and NFSPUFF, CALPUFF, and TSARS+
(PUFF Lagrangian).
High Resolution
Another criterion that can be applied in preferring a dispersion model is the accuracy
of the dispersion model. Accuracy of a dispersion model can be seen by how
accurate the model presents the particle transport in high resolution shown by the
accurate analytical approach in calculating the trajectory. A dispersion model with
high resolution in space and time will be preferred because this study is focused on
understanding the impacts of biomass burning emissions in short distance ranges
from the source.
Dispersion models based on Eulerian and Lagrangian approaches present good
spatial resolution. Eulerian models are lacking in time resolution; whilst Lagrangian
models are excellent in time resolution. Lagrangian models present an accurate
tracking of a particle movement along trajectories as a function of time. The
accuracy of different trajectories for the Lagrangian approach was reviewed to obtain
82
a high resolution in time (Stohl, 1998). The PUFF Lagrangian model is accurate in
calculating the concentrations along trajectories. The available dispersion models for
biomass burning that fulfil this criterion are NFSPUFF, CALPUFF, and TSARS+.
Brisbane Terrain
Brisbane is located at latitude -27.5o and longitude 153o. It has coastal and hilly
areas. The terrain of Brisbane varies from flat to sloping hill terrain. A dispersion
model for complex terrain will be preferred. The dispersion models NFSPUFF,
CALPUFF, TSARS+, and ATHAM are suitable for this study based on the criterion.
Based on these criteria, the available dispersion models are subjectively marked from
1 (least appropriate) to 5 (most appropriate) for this study.
Table 2.3. The characteristics of the dispersion model ranked by these criteria Model Objective
of Study Physical Process
Completeness Consistency Resolution Brisbane location and terrain
Total
SASEM 4 4 4 4 4 3 23 VSMOKE-GIS
4 4 4 4 4 3 23
NFSPUFF 4 4 3 5 5 5 26 CALPUFF 4 4 3 5 5 5 26 TSARS+ 4 4 4 5 5 5 25 ATHAM 4 4 3 3 3 5 22 VALBOX 4 4 4 2 2 2 18
2.8.5. Summary
• Understanding of the fundamental concepts of dispersion modelling is the
first step that should be taken before studying transport mechanisms of
particles related to their impacts. Every dispersion model is derived from
different basic concepts, analytical approaches, assumptions, and
83
applications. A dispersion model is basically classified into two classes by
theoretical concept: Lagrangian and Eulerian models. Based on different
analytical approaches and assumptions, the dispersion models have been
developed to become Gaussian Lagrangian models (Gaussian model), PUFF
Lagrangian models (PUFF model), Gaussian PUFF Lagrangian models
(Gaussian PUFF model), Lagrangian particle models (LPM), Eulerian Box
models (Box model), and coupled Lagrangian Eulerian models (LEM).
• Dispersion modeling for biomass burning is very complicated and involves a
strength source model and a meteorological model as inputs. The complexity
of a dispersion model for biomass burning is also because of the complex
processes involved during the burning process and in emission transport.
There are many factors and parameters playing important roles in these
processes.
• The choice of a dispersion model may be based on the set up criteria. The
criteria are expected to aid in selecting a dispersion model for this study.
Based on the criteria, the available dispersion models for biomass burning
have been ranked. More critical analysis will be needed to decide which
dispersion model is appropriate for this study.
2.9. Conclusions
Biomass burning has become a serious problem regarded as a major contributor of
particle emissions in the atmosphere causing serious impacts on human health and
global changes. Knowledge of particle characteristics, particle production, and
behaviour of particles in the atmosphere are very complex issues that have drawn the
84
attention of researchers to study biomass burning in the last decade. Investigation of
the characteristics of biomass burning particles at its sources and in the atmosphere
has become a priority research topic. Impacts of biomass burning particle emissions
are also significant topics that have been investigated for many years.
Several factors influencing the characteristics of particles from the source and the
properties of particles in the atmosphere are issues investigated in previous studies.
Variability of fuels, moisture content, and burning phase have been identified as
factors influencing characteristics of particles. Physical, chemical, and
thermodynamic processes of particles in the atmosphere, weather conditions, and
atmospheric conditions are other factors considered to affect particle properties in the
atmosphere. Physical properties of particles (mass, number, and size distribution)
from fresh smokes and aged smokes have been measured in laboratories and fields to
get a better understanding of the particle characteristics. Laboratory measurements
have been conducted to investigate factors influencing the properties of particles and
to characterize particles during the burning process. Field measurements have been
conducted to characterize particles from fresh smokes and aged smokes from
different regions around the world.
The composition of biomass burning particles has been studied for many years.
Biomass burning particles mostly contain organic carbons, black carbons, and
inorganic elements. Ratios of those compounds in biomass burning particles have
been reported in a variety of ranges depending on several variables. Burning phase
has been known as a variable to contribute the ratio. Natural conditions mostly
85
contribute to this variation. Characterization of biomass burning particle compounds
is valuable knowledge in understanding their impacts.
Measurements of particulate matter from biomass burning have been carried out for
fresh smokes and aged smokes to characterize the particle property in the sources and
the atmosphere. The method, place and time of the measurements are aimed at
characterisation purposes. For physical characteristics, the previous measurements of
biomass particles were conducted in terms of mass concentration and size
distributions. The measurements were performed in laboratories and in the field.
Field measurements were carried out in situ or airborne.
Australia has serious problems with biomass burning occurring in most parts of the
continent from year to year. Among the States, the Northern Territory and
Queensland are known to suffer from biomass burning every year. Huge areas are
burnt annually. This produces particle emissions in large amounts in the atmosphere
every year.
Biomass burning has serious effects on human health. Epidemiologic and toxicologic
studies have shown the effects of biomass burning emissions, particulate matter and
gasses on human health in terms of morbidity and mortality. Hundreds of biomass
burning emissions have been recognized as toxic and carcinogens that are dangerous
for human health.
86
Knowledge of characteristics of particle size distribution and emission factors from
biomass burning is needed to get a better understanding of the impact assessments.
Quantifying the impact assessments can be conducted by modelling biomass burning
particles emitted from the sources, transported and dispersed in the air. A dispersion
model for biomass burning is very complicated, involving a strength source model
used for modelling particle production, a meteorological model, and a dispersion
model. Data on particle emission factors are vital data as a primary input of a
dispersion model for biomass burning.
2.10. Knowledge Gaps in regards to particle characteristics from biomass
burning in general and in Australia especially.
Characteristics of biomass burning particles have been studied for many years.
Nevertheless, the complexity of factors in particle production, such as variety of
vegetation, moisture content of biomass, burning process, burning rate, mechanisms
of particle production, and characteristics of particles, are still unclear issues in our
understanding of biomass burning in general. The complexity of natural conditions,
uncontrolled burning, and atmospheric conditions add difficulty to obtaining
comprehensive understanding of the characteristics of biomass burning particles.
Laboratory characterization of biomass burning particles under conditions that
closely simulate real situations in the field is a difficult task due to the complexity of
field conditions and has remained a challenge up to now.
Previous studies have recognized the factors influencing characteristics of biomass
burning particles, such as a variety of vegetation and moisture content. However, the
87
factors are not well understood, due to a complexity of natural conditions in fields. In
fact, a diversity of vegetation is found in fields. Every biomass consists of a variable
composition of compounds, including cellulose, hemicelluloses, lignin, and proteins
as a result of photosynthesis process; and a complexity of burning processes
involving physical and chemical reactions, and heat and mass transfers.
Consequently the relationships between those factors and characteristics of biomass
burning particles are not yet well understood.
Burning phase is another issue known as a factor influencing characteristics of
biomass burning particles. Every burning phase has a certain physical condition. At
the beginning of the burning process or ignition phase, oxidation of biomass starts
with a raising of temperature. During the flaming phase, the provided energy is used
to evaporate the water in the biomass cells and decompose the biomass compounds
to become burning products. Moisture content has an important role in completeness
of burning. Less moisture content causes incomplete combustion, which has an
impact on the characteristics of particles. The last burning phase is when oxidation
and transformation processes continue and the availability of heat affect particle
production. However quantitative knowledge of the relationships between these
factors and particle characteristics is still very limited. The effects of the combination
of the two in relation to characteristics of biomass burning particles, still requires
further investigation. Furthermore, the effects of amount of oxygen and rate of
oxygen supply during burning processes on the characteristics of particles are poorly
understood. Consequently, the relationship between the characteristics of particles
and every phase of burning is still unclear.
88
Characteristics of biomass burning particles in terms of size distribution and
emission factor are important to quantify the impact assessments of biomass burning.
For health impacts, particle size distribution is related to the depth of penetration into
the lung. Smaller particles penetrate deeper into the lung. Particle size also has an
impact on the kind of disease as the consequence. Particle emission factor is
correlated to the dose for human exposures. Particle size and emission factors are
also related to particle properties in the atmosphere. Several physical processes of
particles in the atmosphere are linked to their size and emission factor.
Particles of biomass burning plumes have previously been reported with varying
particle size distributions. The particles were reported as experiencing growth during
transport in the atmosphere and increasing in size with age. Biomass burning
particles found in aged smoke were generally in the presence of secondary particles,
containing organic acids, which enriched the process of particle growth. However
there are a number of issues surrounding biomass burning plumes, such as physical
and chemical characteristic of particles, growth mechanisms, factors and processes
influencing particle growth, and processes occurring during the transport in the
atmosphere; that remain poorly understood.
Characteristics of biomass burning particles have been studied in several regions
around the world. Most of the studies were carried out in Africa, North America,
South America, and the Mediterranean. The studies reported that the particle
characteristics varied for different regions. Differences in natural conditions, variety
of fuels, moisture content of fuels, and weather conditions may be factors
89
contributing to the variation of the characteristics of biomass burning particles.
However these factors’ influence on the differences in particle characteristics are still
unclear and speculative.
In fact, Australia suffers from biomass burning every year, contributing a significant
amount of particles to the atmosphere. However, the characteristics of particles
emitted by biomass burning in the states of Australia are poorly known due to a
limited investigation carried out in the regions. Previous campaigns were focused on
characterizing the biomass burning plumes in both the Northern Territory, Australia,
and in parts of Indonesia (Borneo) with the main aim of measurements of the
scattering coefficients, enabling differences to be highlighted between the two
regions. As a result, data on particle size distribution and emission factor from
biomass burning in the Northern Territory of Australia are still limited, and such data
in most states of Australia are unavailable.
90
2.11. References:
Abel, S., J. M. Haywood, J. Li and P. R. Buseck,(2003), Evolution of biomass
burning aerosol properties from an agricultural fire in south Africa.
Geophysical Research Letters, 30(15),1783, doi:10.1029/2003GL017342.
ABS,(2004). Environment bushfires, Australian Bureau of Statistics, Accessed
March 2005, http://www.abs.gov.au/ausstats.
Ahuja, D. R., V. Joshi, K. R. Smith and C. Venkataraman,(1987), Thermal
performance and emission characteristics of unvented biomass burning
cookstoves: a proposed standard method for evaluation. Biomass, 12(4),247-
270.
Albini, F. A., J. K. Brown, E. D. Reinhardt and R. Ottmar,(1995), Calibration of a
large fuel burnout model. International Journal Wildland Fire, 5(3),173-192.
Albini, F. A. and E. D. Reinhardt,(1995), Modelling ignition and burning rate of
large woody natural fuels., International Journal Wildland Fire, 5(2),81-91.
Albini, F. A. and E. D. Reinhardt,(1997), Improved calibration of a large fuel
burnout model. International Journal Wildland Fire, 7(2),21-28.
Anderson, B. E., W. B. Grant, G. L. Gregory, E. V. Browell, J. E. Collins Jr, D. W.
Sachse, D. R. Bagwell, C. Hudgins, D. R. Blake and N. J. Blake,(1996),
Aerosols from biomass burning over the tropical South Atlantic region:
Distributions and impacts. Journal Geophysical Research, 101,24117-24137.
Anderson, B. E., W. B. Grant, G. L. Gregory, E. V. Browell, J. E. Collins Jr, D. W.
Sachse, D. R. Bagwell, C. H. Hudgins, D. R. Blake and N. J. Blake,(1996),
Aerosols from biomass burning over the tropical South Atlantic region:
91
Distributions and impacts. Journal of Geophysical Research, 101,24117-
24137.
Anderson, N. D., R; ,(2002), Airborne reduced nitrogen: ammonia emissions from
agriculture and other resources. Environment International, 1011,1 - 10.
Andreae, M. O., T. W. Andreae, H. Annegarn, J. Beer, H. Cachier, P. Le Canut, W.
Elbert, W. Maenhaut, I. Salma, F. G. Wienhold and T. Zenker,(1998),
Airborne studies of aerosol emissions from savannas fires in southern Africa:
2. Aerosol chemical composition. Journal Geophysical Research, 103,32119-
32128.
Andreae, M. O., E. V. Browell, M. Garstang, G. L. Gregory, R. C. Harris, G. F. Hill,
D. J. Jacob, M. C. Pereira, G. W. Sachse, A. W. Setzer, P. L. Silvia Dias, R.
W. Talbot, A. L. Torres and S. C. Wolfsy,(1988), Biomass burning emissions
and associated haze laser over Amazonia. Journal of Geophysical Research,
93,1509-1527.
Aneja, V. P., P. A. Roelle, G. C. Murray, J. Southerland, J. W. Erisman, D. Fowler,
W. A. H. Asman and N. Patni,(2001), Atmospheric nitrogen compounds II:
emissions, transport, transformation, deposition and assessment. Atmospheric
Environment, 35(11),1903-1911.
Arbex, M. A.,(2000), Assessment of the effects of sugar cane plantation burning on
hospital respiratory admissions. Journal of Air and Waste Management
Association, 50(10),1745-1749.
Arcos, J. C. and M. G. Argus,(1975), Chemical induction of Cancer. Structural Basis
and Biological Mecanisms, vol. IIA. Academic Press.New York.
92
Areskoug, H., P. Camner, S. E. Dahlén, L. Låstbom, F. Nyberg, G. E. Pershagen and
A. Sydbom,(2000), Particles in ambient air - a health risk assessment.
Scandinavian Journal of Work, Environment and Health, 26,1-96.
Ayers, G. P., M. D. Keywood, J. L. Gras, D. Cohen, D. Garton and G. M. Bailey
(1999). Chemical and Physical Properties of Australian Fine Particles : A
Pilot Study. Environment Australia. Canberra, Australia.
Baldasano, J. M., E. Valera and P. Jimenez,(2003), Air quality data from large cities.
The Science of The Total Environment, 307(1-3),141-165.
Bartlett, M. S.,(1956), An Introduction to Stochastic Processes. Cambridge
Univ.Press. London.
Becker, A., E. Schaller and K. Keuler,(2001), Continuous four-dimensional source
attribution for the Berlin area during two days in July 1994. Part II:
Sensitivity studies. Atmospheric Environment, 35(32),5509-5523.
Becker, A., E. Schaller and K. Keuler,(2002), Erratum to "Continuous four-
dimensional source attribution for the Berlin area during two days in July
1994. Part I: The new Euler-Lagrange-model system LaMM5":
[Atmospheric Environment 35 (32) (2001) 5497]. Atmospheric Environment,
36(24),4001-4013.
Bobak, M. and D. A. Leon,(1999), The effect of air pollution on infant mortality
appears spedific for respiratory causes in the postneonatal period.
Epidemiology, 10,666-670.
Boybeyi, Z. and S. Raman,(1995), Simulation of elevated long-range plume transport
using a mesoscale meteorological model. Atmospheric Environment,
29(16),2099-2111.
93
Boybeyi, Z., S. Raman and P. Zannetti,(1995), Numerical investigation of possible
role of local meteorology in Bhopal gas accident. Atmospheric Environment,
29(4),479-496.
Breyfogle, S. and S. Ferguson,(2003). User assessment of smoke dispersion model
for biomass burning, http://fire.rg.fws.gov/ifcc/research/smokemodels.htm.
Brill, S. and D. Evely (1994). Identifying and harvesting edible and medicinal plants.
New York, Harvest Books.
CA EPA (2001). Safe drinking water and toxic enforcement act of 1986, prop.65,
Chemical known to the state to cause cancer and reproductive toxidity, Office
of Environmental Health Hazard Assesment.
Cachier, H., C. Liousse, P. Buartmenard and A. Gaudichet,(1995), Particulate
content of savanna fire emissions. Journal Atmospheric Chemistry, 22,123-
148.
Cahoon, D. R., B. J. Stock, J. S. Levine, W. R. Cofer III and C. C. Chung,(1992),
Evaluation of a technique for satellite-derived estimation of biomass burning.
Journal of Geophysical Research, 97(D4),3805-3814.
Calori, G. and G. R. Carmichael,(1999), An urban trajectory model for sulphur in
Asian megacities: model concepts and preliminary application. Atmospheric
Environment, 33(19),3109-3117.
Campbell, A., Ed.2003). Learning to live with fire. Australia Burning: fire ecology,
policy and management issues. Melbourne, CSIRO Publishing.
Carvalho, J. C., D. Anfossi, S. Trini Castelli and G. A. Degrazia,(2002), Application
of a model system for the study of transport and diffusion in complex terrain
to the TRACT experiment. Atmospheric Environment, 36(7),1147-1161.
94
Chahed, J., V. Roig and L. Masbernat,(2003), Eulerian-Eulerian two-fluid model for
turbulent gas-liquid bubbly flows. International Journal of Multiphase Flow,
29(1),23-49.
Chalupa, D. C., P. E. Morrow, G. Oberdorster, M. J. Utell and M. W.
Frampton,(2004), Ultrafine particle deposition in subjects with asthma.
Environmental Health Perspective, 112,879-82.
Chan, Y. C., R. W. Simpson, G. H. Mctainsh, P. D. Vowles, D. D. Cohen and G. M.
Bailey,(1999), Source apportionment of PM2.5 and PM10 aerosols in
Brisbane (Australia) by receptor modelling. Atmospheric Environment,
33(19),3251-3268.
Chandrasekhar, S.,(1943). Rev. Mod. Phys, 15,1.
Chock, D. P.,(1991), A comparison of numerical methods for solving the advection
equation--III. Atmospheric Environment. Part A. General Topics, 25(5-
6),853-871.
Christensen, J. H.,(1997), The Danish Eulerian hemispheric model -- A three-
dimensional air pollution model used for the arctic. Atmospheric
Environment, 31(24),4169-4191.
Cogan, J. L.,(1985), Monte Carlo simulation of buoyant dispersion. Atmospheric
Environment (1967), 19(6),867-878.
Core, J. E. C., J.A; DeCaesar, R.T; Houck, J.E,(1982), Residential wood combustion
study.
Core, J. E. C., J.A; Neulicht, R.M,(1984), Current and projected impacts of
residential wood combustion on Pasific Northwest air quality. Journal Air
Pollution Control Association, 34,138 - 143.
95
Cramer, H.,(1946), Mathematical Methods of Statistics. Prinstone Univ. Press.
Csanady, G. T.,(1980), Turbulent diffusion in the environment. CD Reidel
Publishing Company. Dordrecht. Holland
Davidson, M. J., K. R. Mylne, C. D. Jones, J. C. Phillips, R. J. Perkins, H. F. J. C.
and J. C. R. Hunt,(1995), Plume dispersion through large groups of obstacles-
-a field investigation. Atmospheric Environment, 29(22),3245-3256.
Degrazia, G. A., D. Anfossi, J. C. Carvalho, C. Mangia, T. Tirabassi and H. F.
Campos Velho,(2000), Turbulence parameterisation for PBL dispersion
models in all stability conditions. Atmospheric Environment, 34(21),3575-
3583.
Delfino, R. J., C. Sioutas and S. Malik,(2005), Potential role of ultrfine particles in
associations between airborne particle mass and cardiovascular health.,
Environmental Health Perspective, 113(8),934-946.
Demirbas, A.,(2004), Combustion characteristics of different biomass fuels. Progress
in Energy and Combustion Science, 30(2),219-230.
Dennis, A., M. Fraser, S. Anderson and D. Allen,(2002), Air pollutant emissions
associated with forest, grassland, and agricultural burning in Texas.
Atmospheric Environment, 36(23),3779-3792.
Dentener, F. J. and P. J. Crutzen,(1994), A three dimensional model of the global
ammonia cycle. Journal Atmosphere Chemistry, 19,331-369.
Dhalla, N. S., R. M. Temsah and T. Netticadan,(2000), Role of oxidative stress in
cardiovascular health., J. Hypertens, 18,655-673.
Ditlevsen, O.,(2003), Stochastic models for atmospheric particle dispersion.
Probabilistic Engineering Mechanics, 18(2),97-106.
96
Dockery, D. W., A. Pope, X. Xu, J. Spengler, D, J. H. Ware, M. E. Fay, B. G. Ferris
and F. E. Speizer,(1993), Mortality risk of air pollution: a prospective cohort
study. New England Journal of Medicine, 329,1753-1759.
Dorland,(1994), Dorland's illustrated medical dictionary. 28th Edition, W.B. Sauders
Co, Philadelphia
Du, S.,(2001), A heuristic Lagrangian stochastic particle model of relative diffusion:
model formulation and preliminary results. Atmospheric Environment,
35(9),1597-1607.
Dubovik, O., B. N. Holben, T. F. Eck, A. Smirnov, Y. J. Kauffman, M. D. King, D.
Tanre and I. Slutsker,(2002a), Variability of absorption and optical properties
of key aerosol types observed in worldwide locations. Journal Atmospheric
Science, 59,590-608.
Dubovik, O., B. N. Holben, T. F. Eck, A. Smirnov, Y. J. Kaufman, M. D. King,
Tanre.D. and I. Slutsker,(2002b), Variability of absorption and optical
properties of key aerosol types observed in worldwide locations., Journal
Atmospheric Science, 59,590-608.
Echalar, F., P. Artaxo, J. V. Martin, M. A. Yamasoe and F. Gerab,(1998), Long-term
monitoring of atmosphere aerosol in the Amazon basin: Source identification
and apportionment., Journal of Geophysical Research, 103,31849-31864.
Eck, T. F., B. N. Holben, J. S. Reid, N. T. O'Neill, J. S. Schaller, O. Dubovik, A.
Smirnov and M. A. Yamasoe,(2003), High aerosol optical depth biomass
burning events: a comparison of optical properties for different source
regions. Geophysical Research Letters, 30(24),2293, doi:
10.1029/2003GL018697.
97
Edwards, R. D., K. R. Smith, J. Zhang and Y. Ma,(2003), Models to predict
emissions of health-damaging pollutants and global warming contributions of
residential fuel/stove combinations in China. Chemosphere, 50(2),201-215.
Elguren-Fernandez, A., A. H. Miguel, P. Jaques and C. Sioutas,(2003), Evaluation of
a denuder MOUDI-PUF sampling system to determine the size distribution of
semivolatile polycyclic aromatic hydrocarbons in the atmosphere. Aerosol
Science and Technology, 37,201-209.
Englert, N.,(2004), Fine particles and human health--a review of epidemiological
studies. Toxicology Letters
Proceedings of EUROTOX 2003. The XLI European Congress of Toxicology.
Science for Safety, 149(1-3),235-242.
EPA,(1986), Technical support document for residential wood combustion. EPA
documen 450/4-85-012, Porland, Oregon
Ermak, D. L. and J. S. Nasstrom,(2000), A Lagrangian stochastic diffusion method
for inhomogeneous turbulence. Atmospheric Environment, 34(7),1059-1068.
Etzel, R.,(1999), A research highlights: Air pollution and bronchitis symptoms in
Southern California children with asthma. Environmental Health
Perspectives, 107(9)
Feller, W.,(1957), An Introduction to Probability Theory and its Application. John
Wiley.New York.
Ferek, R. J., J. S. Reid, P. V. Hobbs, D. R. Blake and C. Liousse,(1998), Emission
factors of hydrocarbons, trace gases and particles from biomass burning in
Brazil. Journal of Geophysical Research, 103,32107-32118.
98
Ferge, T., J. Maguhn, K. Hafner, F. Muhlberger, M. Davidovic, R. Warnecke and R.
Zimmermann,(2005), On-line analysis of gas phase composition in the
combustion chamber and particle characteristics during combustion of wood
and waste in a small batch reactor. Environmental Science and Technology,
39(6),1393-1402.
Ferguson, S. A., J. Peterson and A. Acheson,(2001). Automated, real time
predictions of cumulative smoke impacts from prescribed forest and
agricultural fires. 4th Symposium on Fire and Forest Meteorology, Reno,
Nevada, American Meteorology Society.
Ferrero, E., S. T. Castelli and D. Anfossi,(2003), Turbulence fields for atmospheric
dispersion models in horizontally non-homogeneous conditions. Atmospheric
Environment, 37(17),2305-2315.
Fiebig, M., A. Stohl, M. Wendisch, S. Eckhardt and A. Petzold,(2003), Dependence
of solar radioactive forcing of forest fire aerosol on aging and state of
mixture. Atmospheric Chemistry and Physics, 3,881-891.
Fine, P. M., G. R. Cass and B. R. T. Simoneit,(2001), Chemical characterization of
fine particle emissions from fireplace combustion of woods grown in the
Notheastern United States. Environmental Science and Technology, 35,2665-
2675.
Fine, P. M., G. R. Cass and B. R. T. Simoneit,(2002), Chemical characterization of
fine particle emissions from the fireplace combustion of woods grown in the
Southern United States. Environmental Science and Technology, 36,1442-
1451.
99
Finney, M. A.,(1998), FARSITE Fire Area Simulator Version 1.0: Users Guide and
Technical Documentation. System for Environment Management, P.O. Box
8868, Missoula M T 59807.
Flemming, J., E. Reimer and R. Stern,(2001), Long term evaluation of the ozone
forecast by an Eulerian model. Physics and Chemistry of the Earth,Part B:
Hydrology, Oceans and Atmosphere, 26(10),775-779.
Formenti, P., W. Elbert, W. Maenhaut, J. Haywood, S. Osborne and M. O.
Andreae,(2003), Inorganic and carbonaceous aerosols during the Southern
African Regional Science Initiative (SAFARI 2000) experiment: Chemical
characteristics, physical properties, and emission data for smoke from African
biomass burning. Journal of Geophysical Research, 103(D13),8488,
doi:10.1029/2002JD002408.
Formenti, P., T. Reiner, D. Sprung, M. O. Andreae, M. Wendisch, H. Wex, D.
Kindred, K. Dewey, J. Kent, M. Tzortziou, A. Vasaras and C. Zerefos,(2002),
STAAARTE-MED 1998 summer airborne measurements over the Aegean
Sea: 2. Aerosol scattering and absorption, and radiative calculations. Journal
of Geophysical Research, 107,doi:10.1029/2001JD001536.
Franzese, P.,(2003), Lagrangian stochastic modelling of a fluctuating plume in the
convective boundary layer. Atmospheric Environment, 37(12),1691-1701.
Gao, S., D. A. Hegg, P. V. Hobbs, T. W. Kirchstetter, B. I. Magi and M.
Sadilek,(2003), Water soluble organic components in aerosols associated
with savannas fires in southern Africa: identification, evolution, and
distribution., Journal of Geophysical Research, 108(D13),8491, doi:
10.1029/2002JD002324.
100
Ge, S., W. Liu, Z. Bai, T. Wang, S. Qing, T. Zhu and J. Zhang,(2001), The boiler
breiquette coal versus raw coal: 1. Stack gas emissions. Journal of Air and
Waste Management Association, 51,524-533.
Getler, A. W., J. C. Sagebiel and W. A. Dippel,(1998), Measurements of dioxin and
furan emission from heavy duty diesel vehicles. Journal of Air and Waste
Management Association, 48,276-278.
Gill, A. M., P. G. Ryan, P. H. R. Moore and M. Gibson,(2000), Fire regimes of world
heritage Kakadu National Park Australia. Australia Ecology, 25,616-625.
Gill, M. and P. H. R. Moore,(2005). Fire Situation in Australia, March 2005,
http://www.fao.org/docrep.
Godish, T.,(1989). Formadehyde- our homes and health, Burning Issues,
http://burningissues.or.
Goldberg, M. S., J. C. I. Bailar, R. T. Burnett, J. R. Brook, R. Tamblyn, Y. Bonvalot,
P. Ernst, K. M. Flegel, R. K. Singh and M.-F. Valois,(2000), Identifying
subgoups of the general population that may be susceptible to short-term
increases in particulate air pollution: A time series study in Montreal,
Quebec. Health Effects Institute, Research Report Number 97
Gouesbet, G. and A. Berlemont,(1998), Eulerian and Lagrangian approaches for
predicting the behaviour of discrete particles in turbulent flows. Progress in
Energy and Combustion Science, 25(2),133-159.
Gouesbet, G. and A. Berlemont,(1999), Eulerian and Lagrangian approach for
predicting the behaviour of discrete particles in turbulence flows. Progress In
Energy and Combustion Science, 25,133-159.
101
Gras, J. L.,(1991), Southern hemisphere tropospheric aerosol microphysics., J.
Geophys. Res ,, 96(D3),5345-5365.
Gras, J. L.,(1999), Some optical properties of smoke aerosol in Indonesia and
Tropical Australia. Geophysical Research Letter, 26(10),1393-1396.
Grell, G. A., J. Dudhia and D. R. Stauffer,(1994), A description of the fifth
generation penn State/ NCAR model (MM5). NCAR Technical Note
NCAR/TN-398+STR, National Center for Atmospheric Research, Boulder,
CO.,122.
Gullett, B. K. and A. Touati,(2003), PCDD/F emissions from forest fire simulations.
Atmospheric Environment, 37(6),803-813.
Guyon, P., G. Frank, M. Welling, D. Chand, P. Artaxo, L. Rizzo, G. Nishioka, O.
Kolle, H. Fritsch, M. A. F. S. Dias, M. Cordova and M. O. Andreae,(2005a),
Airborne measurements of trace gas and aerosol particle emissions from
biomass burning in Amazonia. Atmos. Chem. Phys, 5,2791-2831.
Guyon, P., G. Frank, M. Welling, D. Chand, P. Artaxo, L. Rizzo, G. Nishioka, O.
Kolle, H. Fritsch, M. A. F. Silva Dias, L. V. Gatti, M. Cordova and M. O.
Andreae,(2005b), Airborne measurements of trace gas and aerosol emissions
from biomass burning in Amazonia. Atmospheric Chemistry and Physics
Discussion, 5,2791-2831.
Hamwood, K. R.,(1992), Large forest plantation fire in Queensland. IFFN, 7,2-3.
Hanna, S. R.,(1979), Some statistics of Lagrangian and Eulerian wind fluctuation.
Journal of Applied Meteorology, 18,518-531.
Harvey, R. G.,(1991), Polycyclic aromatic hydrocarbons: chemistry and
carcinogenicity. Cambridge University Press.Cambridge.
102
Hays, M. D., C. D. Geron, K. J. Linna, N. D. Smith and J. J. Schauer,(2002),
Speciation of gas-phase and fine particle emissions from burning of foliar
fuels. Environmental Science and Technology, 36,2281-2294.
Haywood, J. M., S. R. Osborne, P. N. Francis, A. Keil, P. Formenti, M. O. Andreae
and K. P. H.,(2003), The mean physical and optical properties of regional
haze dominated by biomass burning aerosol measured from the C-130 aircraft
during SAFARI 2000. Journal Geophysical Research, 108,doi:
10.1029/2002JD002226.
He, Z., Kim, Y. J., Ogunjobi, K. O., and Hong, C. S.,(2003), Characteristics of
PM2.5 species and long-range transport of air masses at Taean background
station, South Korea. Atmospheric Environment, 37(2),219-230.
Hedberg, E., A. Kristensson, M. Ohlsson, C. Johansson, P.-A. Johansson, E.
Swietlicki, V. Vesely, U. Wideqvist and R. Westerholm,(2002), Chemical
and physical characterization of emissions from birch wood combustion in a
wood stove. Atmospheric Environment, 36(30),4823-4837.
HEI,(2000). Morbidity and mortality from air pollutant, Burning issues/Clean Air
Revival, Inc, 2002, www.burningissues.org.
Heinz, S.,(1998), Connections between Lagrangian stochastic models and the closure
theory of turbulence for stratified flows. International Journal of Heat and
Fluid Flow, 19(2),193-200.
Heinz, S. and H. v. Dop,(1999), Buoyant plume rise described by a Lagrangian
turbulence model. Atmospheric Environment, 33(13),2031-2043.
103
Hien, P. D., Binh, N.T., Truong, Y., Ngo, N.T., and Sieu, L.N.,(2001), Comparative
receptor modelling study of TSP, PM2 and PM2-10 in Ho Chi Minh City.
Atmospheric Environment, 35,2669 - 2678.
Hildemann, L. M., G. R. Markoski and G. R. Cass,(1991a), Chemical composition of
emissions from urban sources of fine organic aerosol. Environmental Science
and Technology, 25,744-759.
Hildemann, L. M., G. R. Markoski, M. C. Jones and G. R. Cass,(1991b),
Submicrometer aerosol mass distributions of emissions from boiler, fireplace,
automobiles, diesel trucks, and meat-cooking operations. Environmental
Science and Technology, 14,138-152.
Hinds, W. C.,(1999). Aerosol Technology, 2nd ed. New York. Oxford University
Press
Hobbs, P. V., J. S. Reid, J. A. Herring, J. D. Nance, R. E. Weiss, J. L. Ross, D. A.
Hegg, R. D. Ottmar and C. Liousse,(1996), Particle and gas measurements in
smoke from prescribed burns of forest products in the Pacific Northwest. in:
Biomass Burning and Global Change, Vol. 1. edited by Levine, J.S, pp. 697-
715, MIT Press, New York
Hobbs, P. V., J. S. Reid, J. A. Herring, J. D. Nance, R. E. Weiss, J. L. Ross, D. A.
Hegg, R. D. Ottmar and C. Liousse,(1996), Particle and trace gas
measurements in smoke from prescribed burns of forest products in the
Pacific Northwest , in: Biomass Burning and Global Change, Vol 1,edited by
:Levine, J.S.618-636, MIT Pres, New York.
Holmes, J., R. Atkiston, J. Arey, A. Winer and B. Zielinska (1989). Ambient
concentration of Polycyclic aromatic Hydrocarbons (PAHs) at selected
104
location in California. California Air Resources, Research Report, National
Technical information Service, 5285 Port Royal Rd., Springfield, VA 22161.
Hornig, J. F. S., R.H; barefoot III, C.A; Galasyn, J.F, Ed.1985). Wood smoke
analysis: Vaporization losses of PAH from filter and levoglucosan as a
distinctive make for wood smoke. Polynuclear Aromatic Hydrocarbons:
Mechanisms, Methods, and Metabolism. Colombus, Battlelle Press.
Hueglin, C. H., C. H. Gaegauf, S. Kunzel and H. Burtscher,(1997), Characterization
of wood combustion particles: morphology, mobility, and photoelectric
activity. Environmental Science and Technology, 31,3439-3447.
Hummel, J. and J. Rafsnider,(1995), TSAR plus smoke production and dispersion
model user's guide., Preliminary Draft 7.95. National Biological Service and
the Interior Fire Coordination Committee. Un published report. On file with:
Environmental Science and Technology Center (ESTC), 2401 Research Blvd.,
Suite 205. Fort Collins, CO 80526,p96.
Hurley, P., P. Manins, S. Lee, R. Boyle, Y. L. Ng and P. Dewundege,(2003), Year-
long, high-resolution, urban airshed modelling: verification of TAPM
predictions of smog and particles in Melbourne, Australia. Atmospheric
Environment, 37( 14),1899-1910.
IARC,(1989), World Health Organization, International Agency for Researchon
Cancer. IARC Monographs.Lyon, France. 46, 41-155.
Iliopoulos, I., Y. Mito and T. J. Hanratty,(2003), A stochastic model for solid particle
dispersion in a nonhomogeneous turbulent field. International Journal of
Multiphase Flow, 29(3),375-394.
105
Jacobs, J.,(1998), Rice burning and asthma hospitalizations, Buttle County,
California, 1983-1992. Environmental Health Perspectives, 105(9),980-985.
Jenkins, B. M., R. R. Baker and J. B. Wei,(1996a), On the properties of washed
straw. Biomass and Bioenergy, 10,177 - 200.
Jenkins, B. M., L. L. Baxter, T. R. Miles Jr. and T. R. Miles,(1998), Combustion
properties of biomass. Fuel Processing Technology, 54(1-3),17-46.
Jennings, A. A. and S. J. Kuhlman,(1997), An air pollution transport teaching
module based on GAUSSIAN MODEL 1.1. Environmental modelling &
software, 12,151-160.
Jung, Y.-R., W.-G. Park and O.-H. Park,(2003), Pollution dispersion analysis using
the puff model with numerical flow field data. Mechanics Research
Communications, 30(4),277-286.
Karl, T., A. Hansel, T. Mark, W. Lindinger and D. Hoffmann,(2003), Trace gas
monitoring at the Mauna Loa baseline observatory using proton-transfer
reaction mass spectrometry. International Journal of Mass spectrometry,
233,527-538.
Kauffman, J. B., M. D. Steele, D. L. Cummings and V. J. Jaramillo,(2003), Biomass
dynamics associated with deforestation, fire, and, conversion to cattle pasture
in a Mexican tropical dry forest. Forest Ecology and Management, 176(1-
3),1-12.
Kavouras, I. G., N. Stratigakis and E. G. Stephanou,(1998), Iso and anteiso-alkanes:
specific tracers of environmental tobacco smoke indoor and outdoor particle-
size distributed urban aerosol. Environmental Science and Technology,
32,1369-1377.
106
Khalil, M. A. K. and R. A. Rasmussen,(2002), Tracers of wood smoke. Atmospheric
Environment, In Press, Corrected Proof
Khalil, M. A. K. and R. A. Rasmussen,(2003), Tracers of wood smoke. Atmospheric
Environment, In press
Kim, E. and T. Larson,(2001), Simulation of large particle transport near the surface
under stable conditions: comparison with the Hanford tracer experiments.
Atmospheric Environment, 35(20),3509-3519.
Kim, S., S. Shen and C. Sioutas,(2002), Size distribution and diurnal and seasonal
trends of ultrafine particles in source and receptor sides of Los Angeles
basin., J. Air Waste Manage Assoc, 52,297-307.
Kittelson, D. B.,(1998), Engines and nanoarticles: a review. J. Aerosol Science,
29,575-588.
Kleeman, M. J., J. J. Schauer and G. R. Cass,(1999), Size and composition
distribution of fine particulate matter emitted from wood burning, meat
charbroiling and cigarettes. Environmental Science and Technology, 33,3516-
3523.
Koe, L. C. C., A. F. ArellanoJr. and J. L. McGregor,(2001), Investigating the haze
transport from 1997 biomass burning in Southeast Asia: its impact upon
Singapore. Atmospheric Environment, 35(15),2723-2734.
Koppmann, R., K. V. Craplewski and J. S. Reid,(2005), A review of biomass burning
emissions, part I: gasesous emissions of carbon monoxide, methane, volatile
organic compounds, and nitrogen containing compounds. Atmospheric
Chemistry and Physics Discussion, 5,10455-10516.
107
Korenaga, T., X. Liu and Z. Huang,(2001), The influence of moisture content on
polycyclic aromatic hydrocarbons emission during rice straw burning.
Chemosphere - Global Change Science, 3(1),117-122.
Krejci, R., J. Strom, M. Reus, de, J. Williams, H. Fischer, M. O. Andreae and H. C.
Hansson,(2005), Spatial and temporal distribution of atmospheric aerosols in
the lowermost troposphere over the Amazonian tropical rain forest.,
Atmospheric Chemistry and Physics, 5,1527-1543.
Kulmala, M. and A. Laakssonen,(1990), Binary nucleation of water-sufuric acid
system: comparation of classical theories with different H2SO4 saturation
vapour pressures. Journal Chemistry and Physics, 93,696-701.
Kulmala, M., L. Pirjola and J. M. Makela,(2000), Stable sulphate clusters as a source
of new atmospheric particles. Nature, 404,66-69.
Kulmala, M., H. Vehkamaki, T. Petaja, M. Dal Maso, A. Lauri and V. M.
Kerminen,(2004), Formation and growth rate ultrafine atmospheric particles:
a review of observations. J. Aerosol Science, 35,143-176.
Larson, T. and J. Koenig,(1993), A summary of the emissions characterization and
noncancer repiratory effects of wood smoke. EPA-453/R-93-046-US EPA
Le Canut, P., M. O. Andreae, G. M. Harris, F. G. Weinhold and T. Zenker,(1996),
Airborne studies of emissions from savannas fire in southern Africa, 1,
Aerosol emissions measured with a laser optical particle counter. Journal of
Geophysical Research, 101,23615-23630.
Levine, J.,(1990). Convener of Chapman Conference on Global Biomass Burning
march 19-23, Williamburg, Virginia.
108
Levy, J. I., J. K. Hammit and J. Spengler, D,(2000), Estimate the mortality impacts
of particulate matter: What can be learned from Between-Study Variability ?,
Environmental Health Perspective, 108,109-117.
Li, N., C. Sioutas, J. R. Froines, A. Cho, C. Misra and A. Nel,(2003), Ultrafine
particle pollutants induce oxidative stress and mitochondrial damage.
Environmental Health Perspective, 111,455-460.
Lines, I. G., D. M. Deaves and W. S. Atkins,(1997), Practical modelling of gas
dispersion in low wind speed conditions, for application in risk assessment.
Journal of Hazardous Materials, 54(3),201-226.
Liousse, C., C. Devaux, F. Dulac and H. Cachier,(1995), Aging of savannah biomass
burning aerosols: Consequences on their optical properties. Journal of
Atmospheric Chemistry, 22,1-17.
Lobert, J. M. and J. Warnatz,(1996), Emission from the combustion process in
vegetation, In: Fire in the Environment: The Ecological, Atmospheric and
Climatic Importance of Vegetation Fires, edited by: Cruzen, P.J and
Goldammer, J.G. John Wiley.New York. pp:15-37.
Luke, R. H. and A. G. McArthur,(1978), Bushfires in Australia. Australian
Government Publishing Service, Canberra.
Malcolm, A. L. and A. J. Manning,(2001), Testing the skill of a Lagrangian
dispersion model at estimating primary and secondary particulates.
Atmospheric Environment, 35(9),1677-1685.
Mangia, C., D. M. Moreira, I. Schipa, G. A. Degrazia, T. Tirabassi and U.
Rizza,(2002), Evaluation of a new eddy diffusivity parameterisation from
109
turbulent Eulerian spectra in different stability conditions. Atmospheric
Environment, 36(1),67-76.
McConnel, R., K. Berhane, F. Gilliland, S. J. London, H. Vora, E. Avol, W. J.
Gauderman, H. G. Margolis, F. Lurmann, D. C. Thomas and J. M.
Peters,(1999), air pollution and bronchitis symptoms in Southern California
Children with asthma. Enviromental Health Perspectives, 107,757-760.
McDonald, J. D., B. Zielinska, E. M. Fujita, J. C. Sagebiel, J. C. Chow and J. G.
Watson,(2000), Fine particle and gaseous emission rates from residential
wood combustion. Environmental Science and Technology, 34,2080-2091.
McGrattan, K. B.,(2003), Smoke Plume Trajectory Modelling. Spill Science &
Technology Bulletin, 8(4),367-372.
Meith, P., S. Unger and A. Sydow,(2003), Short term ozone forecasting with
Eulerian dispersion model.
Miles, T. R. M. J., T.R; Baxter, L.L; Bryers, R.W; Jenskins, B.M; Oden,L.L,(1995),
Alkali deposits found in biomass power plants: a preliminary investigation of
their extent and nature. National Renewable Energy laboratory, Golden Co.
Milne, T. A., Brennan, A.H., Glenn, B.H.,(1990), Sourcebook of methods of analysis
fro biomass and biomass conversion process. Elsevier, Amsterdam.
Mitra, A. P., L. Morawska, C. Sharma and J. Zhang,(2002), Chapter two:
methodologies for characterisation of combustion sources and for
quantification of their emissions. Chemosphere, 49(9),903-922.
Moissette, S., B. Oesterle and P. Boulet,(2001), Temperature fluctuations of discrete
particles in a homogeneous turbulent flow: a Lagrangian model. International
Journal of Heat and Fluid Flow, 22(3),220-226.
110
Morawska, L., Bofinger, N.D., Kosic, L., and Nwankwoala, A.,(1998), Submicron
and supermicron particles from diesel vehicles emissions. Environmental
Science and Technology, 32(14),2033 - 2042.
Morawska, L. and J. (Jim) Zhang,(2002), Combustion sources of particles. 1. Health
relevance and source signatures. Chemosphere, 49(9),1045-1058.
Moschandreas, D. J., J. Watson, P. D'Abreton, J. Scire, T. Zhu, W. Klein and S.
Saksena,(2002), Chapter three: methodology of exposure modeling.
Chemosphere, 49(9),923-946.
Mukherji, S., A. K. Swain and C. Venkataraman,(2002), Comparative mutagenicity
assessment of aerosols in emissions from biofuel combustion. Atmospheric
Environment, 36(36-37),5627-5635.
Muraleedharan, T. R., M. Radojevic, A. Waugh and A. Caruana,(2000), Chemical
characterisation of the haze in Brunei Darussalam during the 1998 episode.
Atmospheric Environment, 34(17),2725-2731.
Nel, A., D. Diaz-Sanchez, D. Ng, T. Hiura and A. Saxon,(1998), Enhancement of
allergic inflammation by the interaction between diesel exhaust particles and
the immune system. J. Allergy Clinical Immunology, 102,539-554.
Nel, A. E., D. Diaz-Sanchez and N. Li,(2001), The role of particulate pollutants in
pulmonary inflammation and asthma: evidence for the involvement of
organic chemicals and oxidative stress., Curr Opin Pulm Med, 7,20-26.
Nemmar, A., P. H. M. Hoet, M. Thomeer, B. Nemery, B. Vanquickenborne and H.
Vanbilloen,(2002), Passage of inhaled particles into the blood circulation in
humans. Circulation, 105,411-414.
111
Nemmar, A., M. F. Hoylaetr, P. H. M. Hoet and B. Nemery,(2004), Possible
mechanisms of cardiovascular effects oh inhaled particles: systemic
translocation and prothrombotic effects. Toxicological Letter, 149,243-253.
Nguyen, K. C., J. A. Noonan, I. E. Galbally and W. L. Physick,(1997), Predictions of
plume dispersion in complex terrain: Eulerian versus Lagrangian models.
Atmospheric Environment, 31(7),947-958.
Nichol, J.,(1997), Bioclimatic impacts of the 1994 smoke haze event in southeast
Asia. Atmospheric Environment, 31(8),1209-1219.
Nolte, C. G., J. J. Schauer, G. R. Cass and B. R. T. Simoneit,(1999), Highly polar
organic compounds present in meat smoke. Environmental Science and
Technology, 33,3313-3316.
Norris, G., S. N. YoungPong, T. V. Larson and J. W. Stout,(1999), An association
between fine particles and asthma emergency department visits for children
in Seatle. Environmental Health Perspectives, 107,489-493.
Oanh, N. T. K., L. B. Reutergardh and N. T. Dung,(1999), Emission of polycyclic
aromatic hydrocarbons and particulate matter from domestic combustion of
selected fuels. Environmental Science and Technology, 33,2703-2709.
Oberdorster, G.,(2001), Pulmonary effects on inhaled ultrafine particles. Int Arch
Occup Environ Health, 74,1-8.
Oberdorster, G., Z. Sharp, V. Atudorei, A. Elder, R. M. Gelein and A. Lunts,(2002),
Extrapulmonary translocation of ultrafine carbon particles following whole
body inhilation exposure of rats. J. Toxicology Environmental Health A,
65,1531-1543.
112
Oberdorster, O. G., R. M. Gelein, J. Ferin and B. Weiss,(1995), Association of
particulate air pollution and acute mortality: involvement of ultrafine
particles?, Inhalation Toxicology, 7,111-124.
Oberhuber, J. M., M. Herzog, H. F. Graf and K. Schwanke,(1998), Volcanic plume
simulation on large scales. Journal of Volcano logy and Geothermal
Research, 87,29-53.
Odman, M. T. and A. G. Russell,(1993), A nonlinear filtering algorithm for multi-
dimensional finite element pollutant advection schemes. Atmospheric
Environment. Part A. General Topics, 27(5),793-799.
Oettl, D., J. Kukkonen, R. A. Almbauer, P. J. Sturm, M. Pohjola and J.
Harkonen,(2001), Evaluation of a Gaussian and a Lagrangian model against a
roadside data set, with emphasis on low wind speed conditions. Atmospheric
Environment, 35(12),2123-2132.
Oros, D. R., and and B. R. T. Simoneit,(2001), Identification and emission factors of
molecular tracers in organic aerosols from biomass burning Part 1. Temperate
climate conifers. Applied Geochemistry, 16(13),1513-1544.
Ortiz de Zarate, I., A. Ezcurra, J. P. Lacaux and P. Van Dinh,(2000), Emission factor
estimates of cereal waste burning in Spain. Atmospheric Environment,
34(19),3183-3193.
Osan, J., B. Alfoldy, S. Torok and R. Van Grieken,(2002), Characterisation of wood
combustion particles using electron probe microanalysis. Atmospheric
Environment, 36(13),2207-2214.
113
Otmmar, R. D., M. F. Burn, J. N. Hall and A. D. Hanson,(1993), CONSUME User's
Guide. Gen. Tech. Rep. PNW-GTR-304. U.S. Department of Agriculture,
Forest Service, Pasific Northwest Research Station, Porland, OR,177.
Owczarz, W. and Z. Zlatev,(2002), Parallel matrix computations in air pollution
modelling. Parallel Computing, 28(2),355-368.
Pagels, J., L. Johansson, M. Hagstrom, M. Bohgard, Tullin and M. Sanati,(2002).
Intercomparation of SMPS, ELPI and APS 3320 during sampling of particles
emitted from a domestic wood pellet burner. Abstracts of the sixth
International Aerosol Conference, Taipei, Taiwan.
Pagels, J., Johansson, L., Hagstrom, M., Bohgard, M., Tullin, and Sanati,M.,(2002).
Intercomparation of SMPS, ELPI and APS 3320 during sampling of particles
emitted from a domestic wood pellet burner. The sixth International Aerosol
Conference, Taipei, Taiwan.
Parham, R. A., and Gray, R.L.,(1984), Formation and structure of wood, In:Rowell,
R (Ed), Chemistry of solid wood. American Chemical Society, 207,3 - 56.
Pasquill, F. and F. B. Smith,(1983), Atmospheric diffusion. Wiley, New York, USA.
Pekkanen, J., K. L. Timonen, J. Ruuskanen, A. Responen and A. Mirme,(1997),
Effects of ultrafine and fine particles in urban air on peak flow espiratory
flow among children with asthmatic symptoms. Environmental Research,
74,24-33.
Peters, A., E. Liu, R. L. Verrier, J. Schwartz, D. R. Gold, M. Mittleman, J. Baliff, J.
A. Oh, G. Allen, K. Monahan and D. W. Dockery,(2000), Air pollution and
incidence of cardiac arrhythmia. Epidemiology, 11(1),11-17.
114
Peters, A., S. Perz, A. Doring, J. Stieber, W. Koenig and H. E. Wichmann,(1999),
Increase in heart rate during an air pollution episode. American Journal of
Epidemiology, 150,1094-1098.
Peters, A., H. E. Wichmann, T. Tuch, J. Heinrich and J. Heyder,(1997), Respiratory
effects are associated with the number of ultra fine particles., American
Journal Respiratory Critical Care Medicine, 155,1376-1383.
Peters, J. M., E. Evol, W. Navidi, S. J. London, W. J. Gauderman, F. Lurmann, W. S.
Linn, H. Margolis, E. Rappaport, J. Hong, Jr and D. C. Thomas,(1999), a
study of twelve southern California communitis with differing levels and
types of Air pollution: I. Prevalence of respiratory morbidity. American
Journal Respiratory Critical Care Medicine, 159,760-767.
Peters, L. K., C. M. Berkowitz, G. R. Carmichael, R. C. Easter, G. Fairweather, S. J.
Ghan, J. M. Hales, L. Ruby Leung, W. R. Pennell and F. A. Potra,(1995), The
current state and future direction of Eulerian models in simulating the
tropospheric chemistry and transport of trace species: a review. Atmospheric
Environment, 29(2),189-222.
Pope, C. A., M. J. Thun, M. M. Namboodri, D. W. Dockery, J. S. Evans, F. E.
Speizer and C. W. Heath,(1995), Particulate air pollution as a prediator of
mortality in a propective study of U.S adults. American Journal of
Respiratory Critical Care Medicine, 95,669-674.
Prasad, V. K., Y. Kant, P. K. Gupta, C. Sharma, A. P. Mitra and K. V. S.
Badarinath,(2001), Biomass and combustion characteristics of secondary
mixed deciduous forests in Eastern Ghats of India. Atmospheric Environment,
35(18),3085-3095.
115
Pritchard, R. J., A. J. Ghio, J. R. Lehmann, D. W. Winsett, J. S. Tepper and P.
Park,(1996), Oxidant generation and lung injury after particulate air pollutant
exposure increase with the concentrations associated metals., Inhalation
Toxicology, 8,457-477.
Pyne, S.,(1984), Introduction to Wildland Fire: Fire Management in the United
States. John Wiley.New York.
Radke, L. F., D. A. Hegg, P. V. Hobbs and J. E. Penner,(1995), Effects of aging on
the smoke from a large forest fire. Atmospheric Research., 38,315-332.
Raza, S. S., R. Avila and J. Cervantes,(2001), A 3-D Lagrangian stochastic model for
the meso-scale atmospheric dispersion applications. Nuclear Engineering and
Design, 208(1),15-28.
Reddy, M. S. and C. Venkataraman,(2002a), Inventory of aerosol and sulphur
dioxide emissions from India. Part II--biomass combustion. Atmospheric
Environment, 36(4),699-712.
Reddy, M. S. and C. Venkataraman,(2002b), Inventory of aerosol and sulphur
dioxide emissions from India: I--Fossil fuel combustion. Atmospheric
Environment, 36(4),677-697.
Reid, J. S., T. F. Eck, S. A. Christopher, P. V. Hobbs and B. R. Holben,(1999), Use
of the Angstrom exponent to estimate the variability of optical and physical
properties of aging smoke particles in Brazil., Journal of Geophysical
Research, 104,27489-27498.
Reid, J. S. and P. V. Hobbs,(1998), Physical and optical properties of smoke from
individual biomass fires in Brazil. Journal of Geophysical Research,
103,32013-32031.
116
Reid, J. S., P. V. Hobbs, R. J. Ferek, D. Blake, J. V. Martins, M. R. Dunlap and C.
Liousse,(1998a), Physical, chemical, and optical properties of regional hazes
dominated by smoke in Brazil. Journal of Geophysical Research, 103,32059-
32080.
Reid, J. S., P. V. Hobbs, A. L. Rangno and D. A. Hegg,(1998b), Relationships
between cloud droplet effective radius, liquid water content, and droplet
concentration for warm clouds in Brazil embedded in biomass smoke.
Journal Geophysical Research, 104,6145-6153.
Reid, J. S., R. Koppmann, T. F. Eck and D. P. Eleuterio,(2005), A review of biomass
burning emissions part II: intensive physical properties of biomass burning
particles. Atmospheric Chemistry and Physics, 5,799-825.
Ristovski, Z., Morawska, L., Bofinger, N.D.,(1998), Submicrometer and
supermicrometer particles from spark ignition vehicle emissions.
Environmental Science and Technology, 345,3845 - 3852.
Rogge, W. F., M. A. Marzuek, L. M. Hildemann, G. R. Cass and B. R. T.
Simoneit,(1991), Source of fine organic aerosol: 1. Charboiler and meat
cooking operations. Environment Science and Technology, 25,1112-1125.
Rogge, W. F., M. A. Marzuek, L. M. Hildemann, G. R. Cass and B. R. T.
Simoneit,(1993a), Source of fine organic aerosol:2. Noncatalyst and catalyst-
equipped automobiles and heavy duty diesel trucks. Environmental Science
and Technology, 27,636-651.
Rogge, W. F., M. A. Marzuek, L. M. Hildemann, G. R. Cass and B. R. T.
Simoneit,(1993b), Source of fine organic aerosol: 3. Road dust, tire dibris,
117
and organometallic brake lining dust: roas as sources and sinks.
Environmental Science and Technology, 27,1892-1904.
Rogge, W. F., M. A. Marzuek, L. M. Hildemann, G. R. Cass and B. R. T.
Simoneit,(1993c), sorce of fine organic aerosol: 5. Natural gas home
appliances. Environmental Science and Technology, 27,2736-2744.
Rogge, W. F., M. A. Marzuek, L. M. Hildemann, G. R. Cass and B. R. T.
Simoneit,(1993d), Qualification of organic aerosl on molecular level:
Identification, abundance, and seasonal variation. Atmospheric Environment,
27A,1545 - 1565.
Rogge, W. F., M. A. Marzuek, L. M. Hildemann, G. R. Cass and B. R. T.
Simoneit,(1994), Source of fine organic aerosol: 6. Cigarette smoke in the
urban atmosphere. Environmental Science and Technology, 28,1375-1388.
Rogge, W. F., M. A. Marzuek, L. M. Hildemann, G. R. Cass and B. R. T.
Simoneit,(1997), Source of fine organic aerosol: 7. Hot asphalt roofing tar pot
fumes. Environement Science and Technology, 31,2726-2730.
Rogge, W. F., M. A. Marzuek, L. M. Hildemann, G. R. Cass and B. R. T.
Simoneit,(1998), Source of fine organic aerosol: 9. Pine, oak, and synthetic
log combustion in residential firepalces. Environmental Science and
Technology, 32,13-22.
Rozenberg, M.,(2002). Burning issue website, Burning issue website, 12 Mei 2003,
http://burningissues.org.
Russell-Smith, J., A. C. Edwards and G. D. Cook,(2003a), Reability of biomass
burning estimates from savannas fires: Biomass burning in northern Australia
during the 1999 biomass burning and lighting experiment B field campaign.
118
Journal of Geophysical Research, 108(D3),8405,
doi:10.1029/2001JD000787.
Russell-Smith, J., C. Yates, A. Edwards, G. E. Allan, G. D. Cook, P. Cooke, R.
Craig, B. Health and R. Smith,(2003b), Contemporary fire regimes of
northern Australia, 1997-2001: change since Aboriginal occupancy,
challenges for sustainable management. International Journal of Wildland
Fire, 12,283-297.
Samet, J. M., F. Dominici, F. C. Curreiro, I. Coursac and S. L. Zeger,(2000c), Fine
particulate air pollution and mortality in 20 US cities. New England Journal
of Medicine, 343(24),1742-1749.
Samet, J. M., F. Dominici, S. L. Zeger, J. Schwartz and D. W. Dockery (2000a). The
national morbidity , mortality, and air pollution study. Part I : Methods and
Mehodologic Issues, Health Effects Institute Research Report 94, Part I.
Samet, J. M., S. L. Zeger, F. Dominici, F. C. Curreiro, I. Coursac, D. W. Dockery, J.
Schwartz and A. Zanobetti (2000b). The national morbidity, mortality, and
air pollution study. Part II: Morbidity, Mortality and Air Pollution in the
United States, Health Effects Institute Research Report 94, Part II.
Sanberg, D. V. and J. Peterson,(1984). A source strength model for prescribed fires
in coniferous logging slash. In : Proceeding of the 21st annual meeting of the
air Pollution Control Association, Portland, OR. Seattle, W.A, U.S.
Department of Agriculture, Forest Service, Forest Residues and Energy
Program.
Schauer, J. J., Kleeman, M.J., Cass, G.R., and Simoneit, B.R.T., ,(2001),
Measurement of emissions from air pollution sources. 3. C1-C29 organic
119
compounds from fireplace combustion of wood. Environmental Science and
Technology, 35(9),1716 - 1728.
Schauer, J. J., M. J. Kleman, G. R. Cass and B. R. T. Simoneit,(1999a),
Measurement of emissions from air pollution sources:1. C1 through C29
organic compounds from meat charboiling. Environement Science and
Technology, 33,1566-1577.
Schauer, J. J., M. J. Kleman, G. R. Cass and B. R. T. Simoneit,(1999b),
Measurement of emissions from air pollution sources:1. C1 through C30
organic compounds from medium duty diesel trucks. Environement Science
and Technology, 33,1578-1587.
Schauer, J. J., M. J. Kleman, G. R. Cass and B. R. T. Simoneit,(2001), Measurement
of emissions from air pollution sources:3. C1 through C29 organic
compoundsfire place combustion of wood. Environmental Science and
Technology, 35(9),1716-1728.
Schultz, T. P., and Taylor, F.W.,(1989), Wood, In: O.Kitani, C.W.Hall (Ed). Gordon
& Breach.New York.
Schwartz, J.,(1993), Particulate air pollutant and chronic respiratory disease.
Environmental Research, 62,7-13.
Scire, J., D. G. Strimaitis, R. J. Yamartino and X. Zhang,(1995), A user's guide for
CalPuff dispersion model. sigma Research/ Earth Tech. Concord,
Massachusettes,Doc # 1321-2, 315.
Seaton, A., A. Soutar, V. Crawford, R. Elton, S. McNerian and J. Cherrie,(1999),
Particulate air pollution and the blood. Thorax, 54,1027:1032.
120
Seiler, W. and P. J. Crutzen,(1980), Estimates of gross and net fluxes of carbon
between the biosphere and the atmosphere from biomass burning. Climatic
Change, 14,243-262.
Seland, O. and T. Iversen,(1999), A scheme for black carbon and sulphate aerosols
tested in a hemispheric scale, Eulerian dispersion model. Atmospheric
Environment, 33(17),2853-2879.
Sestak, M. L., W. E. Marlatt and A. R. Riebau (1989). VALBOX:ventilated valey
box model, (unpublished report), on file with : Michael Sestak, U.S.
Department of the Interior, Bureau of Land Management, and Colorado State
University, Environmental Science and Technology Center, 2401 Research
Blvd, Suite 205, Fort Collins, Co 80526, p32.
Sestak, M. L. and A. R. Riebau,(1988), SASEM, Simple Approach Smoke
Estimation Model. U.S Bureau of Land management, Technical Note 382, 31.
Shafizadeh, F.,(1984), The chemistry of pyrolysis and combustion. Chemistry of
Solid Wood. In: Rowell, R. E. Advance Chemistry, American Chemical
Society. Washington, DC. 207, 489-529.
Sherwin, T. J.,(1999), LAGCARTW: A random walk particle advection-diffusion
model. Marine Models, 1(1-4),83-102.
Simoneit, B. R. T.,(1998), Biomaker PAHs in the environment. The Handbook of
Environmental Chemistry, Vol 3. part I., In: Neilson, A. E. Springer
Verlag.Berlin. 175-221.
Simoneit, B. R. T.,(2002), Biomass burning -- a review of organic tracers for smoke
from incomplete combustion. Applied Geochemistry, 17(3),129-162.
121
Simoneit, B. R. T. and V. O. Ellias,(2001b), Detecting organic tracers from biomass
burning in the atmosphere. Marine Pollution Bulletin, 42(10),805-810.
Simoneit, B. R. T., Rogge, W.F., Mazurek, M.A., Standley, L.J., Hildemann, L.M.,
Cass, G.R.,(1993), Lignin pyrolysis products, lignan and resin acids as
specific tracers of plant classes in emissions from biomass combustion.
Environment Science and Technology, 27,2533 - 2541.
Simoneit, B. R. T., Schauer, J.J., Nolte, C.G., Oros, R.R., Elias, V.O., Fraser, M.P.,
Rogge, W.F., Cass, G.R.,(1999), Levoglucosan a tracer for cellulose in
iomass burning and atmospheric particles. Applied Geochemistry, 17(129 -
162.)
Skaar, C.,(1984), Wood-water relationships. Chemistry of solid wood. In: Rowel, R.
E. Advance Chemistry, American Chemical Society. Washington DC. 207,
127-128.
Smith, F. B.,(1968), Conditioned particle motion in a homogeneous turbulent field.
Atmospheric Environment, 2,491-508.
Souto, M. J., J. A. Souto, V. Perez-Munuzuri, J. J. Casares and J. L.
Bermudez,(2001), A comparison of operational Lagrangian particle and
adaptive puff models for plume dispersion forecasting. Atmospheric
Environment, 35(13),2349-2360.
Sportisse, B.,(2001), Box models versus Eulerian models in air pollution modelling.
Atmospheric Environment, 35(1),173-178.
Stanier, C. O., A. V. Khlystov and S. N. Pandis,(2004), Nucleation events during the
Pittsburgh Air Quality Study: description and relation to key meteorological,
122
gas phase, and aerosol parameters. Aerosol Science and Technology, 38(suppl
1),118-126.
Stohl, A.,(1998), Computation, accuracy and applications of trajectories--A review
and bibliography. Atmospheric Environment, 32(6),947-966.
Taylor, G. I.,(1921), Statistical theory of turbulence. Journal Mathematical and
Physical Sciences, 20,196.
Thomas, S. and L. Morawska,(2002), Size-selected particles in an urban atmosphere
of Brisbane, Australia. Atmospheric Environment, 36(26),4277-4288.
Thomson, D. and M. Montgomery,(1994), Refection boundary conditions for
random walk models of dispersion in non Gaussian turbulence. Atmospheric
Environment, 28(1981-1987)
Todd, J. J.,(1991), Emission and performance of woodheaters when burning
softwood. Report 3, Centre for environment Studies, University of Tasmania,
Hobart
Tolbert, P. E., J. A. Mulholland, D. D. MacIntosh, F. Xu, D. Danniel, O. J. Devine,
B. P. Carlin, M. Klein, J. Dorley, A. J. Butler, D. F. Nordenberg, H. Frunkin,
P. B. Ryan and M. C. White,(2000), Air quality and pediatric emergency
room visit for asthma in Atlanta, Georgia. American Journal of
Epidemiology, 151,798-810.
Tombrou, M., E. Bossioli and D. Lalas,(1998), An application of a simple Monte
Carlo dispersion model in complex terrain. Environmental Modelling and
Software, 13(1),45-58.
Torigoe, K.,(2000), Influence of emission from rice straw burning on bronchia
asthma in children. Paediatrics International, 42,143-150.
123
Trentmann, J., M. O. Andreae, H. F. Graf, P. V. Hobbs, R. D. Ottmar and T.
Trautmann,(2002), Simulation of a biomass burning plume: comparison of
model results with observations. Journal of Geophysical Research, 107,D2.
Tsuang, B.-J.,(2003), Quantification on the source/receptor relationship of primary
pollutants and secondary aerosols by a Gaussian plume trajectory model: Part
I--theory. Atmospheric Environment, 37(28),3981-3991.
Tsutsumi, Y.,(1999), Aircraft measurement of ozone, NOx, CO, and aerosol
concentrations in biomass burning smoke over Indonesia and Australia in
October 1997: Depleted ozone layer at low altitude over Indonesia.,
Geophysical Research Letter, 26(5),595-598.
Uherek, E.,(2004). Vegetation fire, Max Planck Institute for Chemistry, Mainz,
Accessed Mei 2004, http://www.atmosphere.mpg.de/enid/238.html.
Ulke, A. G.,(2000), New turbulent parameterization for a dispersion model in the
atmospheric boundary layer. Atmospheric Environment, 34(7),1029-1042.
van Baten, J. M. and R. Krishna,(2001), Eulerian simulations for determination of
the axial dispersion of liquid and gas phases in bubble columns operating in
the churn-turbulent regime. Chemical Engineering Science, 56(2),503-512.
Vedal, S., J. Petkau, R. White and J. Blair,(1998), Acute affects on ambient inhalable
particles in asthmatic and nonasthmatic children. American Journal
Respiratory Critical Care Medicine, 157(4),1034-1043.
Venkatesan, R., R. Mathiyarasu and K. M. Somayaji,(2002), A study of atmospheric
dispersion of radionuclides at a coastal site using a modified Gaussian model
and a mesoscale sea breeze model. Atmospheric Environment, 36(18),2933-
2942.
124
Von Klot, S., G. Wolke, T. Tuch, J. Heinrich, J. Heyder, H. E. Wichmann and A.
Peters,(2000), Short term effects of ultrafine and fine particles on medication
use in asthmatic adults. Proc. Conf. American Thoracic Soc.,2000. Toronto
(abstract).
Warneck, P.,(1988), Chemistry of The Natural Atmosphere. Academic Press. New
York.
Weil, J. C.,(1990), A diagnosis the asymmetry in top-down and bottom-up diffusion
using a Lagrangian stochastic model. Journal of Atmospheric Science,
47(501-517)
WHO,(1999), World Health Organization: Health Guidelines for Vegetation Fire
Events. World Health Organization. Geneva, Switzerland.
WHO,(2000), Vegetation Fires. Http://www.who.int/mediacentre/factsheets/fs254/
en/print.html
WHO,(2006). WHO challenges world to improve air quality: Stricter air pollution
standards could reduce deaths in polluted cities by 15 %, Accessed october
2006, http://www.who.int/mediacentre/news/releases/2006/pr52/en/
print.html.
Wichmann, H. E. and A. Peters,(2000), Epidemiological evidence of the effects of
ultrafine particle exposure. Phil. Trans. R. Soc. Lond. A, 358,2751-2769.
Wichmann, H. E., C. Spix, T. Tuch, G. Wolke, A. Peters and J. Heinrich,(2000),
Dailiy mortality and fine and ultrafine particles in Erfurt, Germany. Part i:
Role of particle number and particle mass. Res Rep Health Eff Inst, 98,5-86.
Wieser, U. and C. k. Gaegauf, (2005). Nanoparticle emissions of wood combustion
processes, Laboratories for Sustainable Energy System, Accessed March
125
2005, http://www.oekozentrum.ch/downloads/publikationen/
nanoparticles.pdf.
Williams, R. J., A. M. Gill and P. H. R. Moore,(1998), Seasonal changes in fire
behaviour in a tropical savannas in northern Australia. International Journal
Wildland Fire, 8,227-239.
Winebrake, J. J. and M. L. Deaton, (1999), Hazardous air pollution from mobile
sources: a comparation of alternative fuel and reformulated gasoline vehicles.
Journal of Air and Waste Management Association, 49,576-581.
Woodruff, T. J., J. Grillo and K. C. Schoendrof,(1997), The relationship between
selected causes of postneonatal inflant mortality and particulate air pollution
in the United States. Environmental Health Perspective, 105,607-612.
Wurzler, S. and M. Simmel,(2005). Impact of vegetation fires on composition and
circulation of the atmosphere, Accessed March 2005,
http://projects.tropos.de:8088/afo200g3/.
Yamartino, R. J., J. S. Scire, G. R. Carmichael and Y. S. Chang,(1992), The
CALGRID mesoscale photochemical grid model--I. Model formulation.
Atmospheric Environment. Part A. General Topics, 26(8),1493-1512.
Ye, B., Ji, X., Yang, H., Yao, X., Chan, C.K., Cadle, S.H., Chan, T., end Mulawa,
P.A.,(2003), Concentration and chemical composition of PM2.5 in Shanghai
for a 1-year period. Atmospheric Environment, 37,499 - 510.
Yu, F. and R. P. Turco, (2000), Ultrafine aerosol formation via ion-mediated
nucleation. Geophysical Research Letter, 27,883-886.
126
Yu, O., L. Sheppard, T. Lumley, J. Q. Koenig and G. G. Shapiro,(2000), Effects of
ambient air pollution on symptoms of asthma in Seattle- area children in the
CAMP study. Environmental Health Perspectives, 108(12),1209-1214.
Zannetti, P.,(1984), New Monte Carlo scheme for simulating Lagrangian particle
diffusion with wind shear effects. Applied Mathematical Modelling, 8(3),188-
192.
Zannetti, P.,(1990), Air Pollution modelling. Van Nostrand Reinhold.New York.
Zhang, J., K. R. Smith, Y. Ma, S. Ye, F. Jiang, W. Qi, P. Liu, M. A. K. Khalil, R. A.
Rasmussen and S. A. Thorneloe,(2000), Greenhouse gases and other airborne
pollutants from household stoves in China: a database for emission factors.
Atmospheric Environment, 34(26), 4537-4549.
Zhou, Q. and M. A. Leschziner,(1999), An improved particle-locating algorithm for
Eulerian-Lagrangian computations of two-phase flows in general coordinates.
International Journal of Multiphase Flow, 25(5), 813-825.
Zou, L. Y., and Hopper, M.A., (1997), Size-resolved airborne particles and
morphology in central Jakarta. Atmospheric Environment, 31(8),1167-1172.
Zou, L. Y., W. Zhang and S. Atkiston,(2003), The characterisation of polycyclic
aromatic hydrocarbons emissions from burning of different firewood species
in Australia. Environmental Pollution, 124(2),283-289.
127
128
CHAPTER 3
QUANTIFICATION OF PARTICLE NUMBER AND MASS
EMISSION FACTORS FROM COMBUSTION OF QUEENSLAND
TREES
Arinto Y.P. Wardoyo, Lidia Morawska, Zoran D. Ristovski, and Jack Marsh
International Laboratory for Air Quality and Health Queensland University of Technology, Brisbane, Queensland, Australia
(2006) Environmental Science and Technology 40, 5696-5703
129
STATEMENT OF JOINT AUTHORSHP
Title: Quantification of particle number and mass emission factors from combustion of Queensland trees
Authors: Arinto Y.P. Wardoyo, Lidia Morawska, Zoran D. Ristovski, and Jack Marsh
Arinto Y. P. Wardoyo (candidate) Developed experimental design and scientific method; conducted experiments; analysed and interpreted data; and wrote manuscript Lidia Morawska Contributed to experimental design and scientific method; interpreted data; and assisted with manuscript Zoran D. Ristovski Contributed to experimental design and scientific method and interpreted data Jack Marsh Assisted in identification and collection of samples
130
ABSTRACT
The quantification of particle emission factors under controlled laboratory
conditions for burning of the following five common tree species found in South East
Queensland forests: Spotted Gum (Corymbia citriodora), Blue Gum (Eucalyptus
tereticornis), Bloodwood (Eucalyptus intermedia), Iron Bark (Eucalyptus crebra),
and Stringybark (Eucalyptus umbra) has been studied. The results of the study show
that the particle number emission factors and PM2.5 mass emission factors depend on
the type of tree and the burning rate. For fast burning conditions, the average particle
number emission factors are in the range of 3.3 to 5.7 x 1015 particles/kg for woods
and 0.5 to 6.9 x 1015 particles/kg for leaves and branches, and the PM2.5 emission
factors are in the range of 140 to 210 mg/kg for woods and 450 to 4700 mg/kg for
leaves and branches. For slow burning conditions, the average particle number
emission factors are in the range of 2.8 to 44.8 x 1013 particles/kg for woods and 0.5
to 9.3 x 1013 particles/kg for leaves and branches, and the PM2.5 emissions factors are
in the range of 120 to 480 mg/kg for woods and 3300 to 4900 mg/kg for leaves and
branches.
Keywords: Biomass burning, emission factors, ultrafine particles, particle number emission.
131
CHAPTER 4
BIOMASS BURNING INFLUENCED PARTICLE CHARACTERISTICS IN
NORTHERN TERRITORY AUSTRALIA BASED ON
AIRBORNE MEASUREMENTS
1Zoran D. Ristovski*, 1Arinto Y.P. Wardoyo, 1Lidia Morawska, 2Milan Jamriska, 2S.
Carr and 1Graham Johnson.
1International Laboratory for Air Quality and Health Queensland University of Technology (ILAQH)
2 Defence Science and Technology Organisation (DSTO)
Submitted for publication in Journal of Geophysical Research
161
STATEMENT OF JOINT AUTHORSHP
Title: Biomass burning influenced particle characteristics in Northern Territory of Australia based on airborne measurements. Authors: Zoran D. Ristovski, Arinto Y.P. Wardoyo, Lidia Morawska, Milan Jamriska, 2Steve Carr and 1Graham Johnson
Zoran D. Ristovski Contributed to experimental design and scientific method, and the data collection, the airborne measurements, assisted with manuscript. Arinto Y. P. Wardoyo (candidate) Processed and analyzed data of the airborne measurements; and wrote manuscript. Lidia Morawska Contributed to interpret data and manuscript Milan Jamriska Contributed to the data collection for the airborne measurements Steve Carr Contributed to the data collection for the airborne measurements Graham Johnson Contributed to instrumentation preparation for the airborne measurements
162
Abstract
Airborne measurements of particle number concentrations from biomass burning were
conducted in the Northern Territory, Australia, during June and September campaigns in
2003, which is the early and the late dry season in that region. The airborne
measurements were performed along horizontal flight tracks, at several heights in order
to gain insight into the particle concentration levels and their variation with height lower
boundary layer (LBL), upper boundary layer (UBL), and also in the free troposphere
(FT). The boundary layer (BL) was obtained to be 1800 m and 1950 m for the June and
September campaign respectively. The measurements found that the concentration of
particles during the early dry season was lower than that of the late dry season. For the
June campaign, the concentration of particles in LBL, UBL, and FT was measured (685
± 245) particles/cm3, (365 ± 183) particles/cm3, and (495 ± 45) particle/cm3 respectively.
For the September campaign, the concentration of particles was found (1233 ± 274)
particles/cm3 in LBL, (651 ± 68) particles/cm3 in UBL, and (568 ± 70) particles/cm3 in
FT. It was also concluded that most likely particles from fresh smoke dominated in the
early dry season, while in the late dry season the plumes contained more aged particles.
The decrease in particle concentration above the boundary layer was found to be 47% in
both the early dry season and the late dry season.
Keywords: Biomass burning, particle number concentration, Northern Territory
Australia, airborne measurements.
163
CHAPTER 5
SIZE DISTRIBUTION OF PARTICLES EMITTED FROM GRASS FIRES IN
THE NORTHERN TERRITORY AUSTRALIA
1Arinto Y.P. Wardoyo, 1Lidia Morawska, 1Zoran D. Ristovski, 2Milan Jamriska,
2Steve Carr and 1Graham Johnson.
1International Laboratory for Air Quality and Health Queensland University of Technology (ILAQH),
GPO Box 2434, Brisbane, Queensland 4001, Australia. 2 Defence Science and Technology Organisation (DSTO)
Submitted for publication in Atmospheric Environment
193
STATEMENT OF JOINT AUTHORSHP
Title: Size distribution of particles emitted from grass fires in the Northern Territory, Australia. Authors: Arinto Y.P. Wardoyo, Lidia Morawska, Zoran D. Ristovski, Milan Jamriska, 2Steve Carr and 1Graham Johnson
Arinto Y. P. Wardoyo (candidate) Developed experimental design, scientific method, conducted experiments, analyzed and interpreted data of the laboratory experiments; processed and analyzed data of the airborne measurements; and wrote manuscript. Lidia Morawska Contributed to interpret data; and assisted with manuscript Zoran D. Ristovski Contributed to experimental design and scientific method, and conducted the airborne measurements. Milan Jamriska Conducted the airborne measurements Steve Carr Contributed to the airborne measurements Graham Johnson Contributed to instrumentation preparation for the airborne measurements
194
Abstract
This study presents the results of investigations of particle size distribution
originating from burning of several grass species under controlled laboratory
conditions, and also at several heights during field campaigns conducted during dry
season in the Northern Territory, Australia. The laboratory study simulated, as
accurately as possible, real field conditions, such as burning phases and burning rate.
Most of particles released in the controlled burning were found to be of a diameter
between 30 and 210 nm, depending on the burning conditions. Under fast burning
conditions, smaller particles were produced with a diameter in the range of 30 to 60
nm. Larger particles, with the diameter between 60 nm and 210 nm, were produced
during slow burning. The airborne field measurements of biomass particles found
that most of the particles measured during the early dry season (EDS) came from
fresh smokes and most of the particles measured during the late dry season (LDS)
were from aged smokes. Under the boundary layers, particles with the CMD of (83 ±
13) nm were obtained during the EDS, and particles with the CMD of (127 ± 6) nm
were found during the LDS. Vertical profiles of particle CMD showed that smaller
particles were found in the higher elevations within the atmosphere. These
measurements provide insight into the scientific understanding of the properties of
biomass burning particles in the Northern Territory, Australia.
Keywords : Biomass burning, particle size distribution, Northern Territory of
Australia, airborne measurement, vertical profile.
195
5.1. INTRODUCTION Among many sources of air pollution, biomass burning has been identified as a
major contributor of particles and gasses in the atmosphere. Natural and
anthropogenic types of biomass burning include forest fires, agricultural and waste
burning, logging slash, land clearing slash and burning for cooking and heating,
Their emissions have played a significant role in influencing changes in atmospheric
processes (Bodhaine, 1983; Shaw, 1987); causing the acidification of clouds, rain,
and fog (Nichol, 1997); and impacting on the transport of UV radiation within the
atmosphere (Wurzler and Simmel, 2005).
Knowledge of the characteristics of the emitted particle has been identified as a very
important element in developing a quantitative assessment of the impact of these
fires. In addition to emission factors, which quantify the magnitude of emissions,
understanding of size distribution of the emitted particles is of significance as it is
size, which determines particle dynamics in the air as well as the impact the particles
have on the environment and the receptors. The studies reported in literature showed
that the majority of particles resulting from biomass burning were less than 2.5 μm in
diameter (Hays et al., 2002; Hedberg et al., 2002; Ferge et al., 2005; Wieser and
Gaegauf, 2005), depending on fuel variability, moisture content, burning conditions
and burning processes. However, quantitative knowledge of the relationships
between these factors and particle size distribution is still very limited. Several
laboratory studies reported that particle size was proportional to flame size
(Glasmann, 1988) and fire intensity (Cofer III et al., 1996; Reid and Hobbs, 1998),
196
however the effects of the combination of the two, in relation to particle size
distribution, still requires further investigation.
Particle size distributions, from the smoke produced by biomass burning, have
previously been reported in accumulation mode with a count median diameter
(CMD) in the range of 0.10 to 0.18 μm. Smoke particles were reported to grow
during transport in the atmosphere and to increase in size with age (Reid et al.,
1999b). The growth of particles from biomass burning has been shown to occur
within hours after emission (Abel et al., 2003) or on time scales of days (Radke et al.,
1995; Reid et al., 1999b). During aging of the smoke many species of secondary
particles are formed, including those formed from organic acids (Gao et al., 2003;
Haywood et al., 2003).
Characteristics of biomass burning particles have been studied in several regions of
the world, showing that particle size varied between the regions, however there were
very few studies conducted so far of the characteristics of biomass particles over the
continent of Australia. The first reported study was conducted mainly over the
Eastern part of the continent (Gras, 1991). Later studies investigated smoke
characteristics in both the Northern Territory, Australia and the parts of Indonesia
(Borneo) (Gras, 1999; Tsutsumi, 1999), mainly focusing on measurements of the
scattering coefficient.
This paper presents the results of a study on particle size distribution produced from
burning of grasses dominating the vegetation of the savannah covering the Northern
Territory, Australia. The experimental parts of the study included controlled
laboratory combustion of the grasses, as well as airborne measurements of the
197
particle size distribution over of the Northern Territory which experiences extensive
fires every year. The aims of the study were to develop a better general
understanding of the impact of fuel composition and burning conditions on particle
size distribution and also to provide more insight on particle characteristics emitted
by biomass burning of the large areas of the Northern Territory, Australia. This study
was a part of a larger Australian research project on particle emissions from biomass
burning, involving the International Laboratory for Air Quality and Health,
Queensland University of Technology (ILAQH, QUT), the Defence Science and
Technology Organisation (DSTO) and the Australian Commonwealth Scientific and
Industrial Research Organization (CSIRO). Other elements of the project included
investigations of particle concentration profiles and their relationships with the
meteorological conditions.
5. 2. METHODS
5.2.1 Laboratory Measurements
The measurements of particle size distribution were conducted by burning grass
samples collected from the savannah of the Northern Territory, Australia, under
controlled laboratory conditions designed to simulate as close as possible the
processes of fast and slow burning occurring in real life. The measurements were
conducted by burning the grass samples in a stove, and then diluting the sampled
smoke in two steps, firstly with compressed fresh air from an ejector dilutor and
secondly by mixing it with filtered air. Particle number emission factors were
quantified by measuring the total particle concentration during the combustion
198
process using a Condensation Particle Counter (CPC), while the size distribution of
particles was measured using a Scanning Mobility Particle Sizer (SMPS).
5.2.1.1 Experimental Setup
The experimental setup consisted of a burning system (modified stove), a dilution
and sampling system and a particle measurement system (Wardoyo et al., 2006).
A modified commercial stove, with the dimensions of 66 x 74.5 x 55 cm3, was
used to simulate burning rates that would occur in field. The part of the stove
originally used to adjust air flow, was replaced by a ventilation system that
enabled the introduction of a controlled amount of air into the stove. In order to
obtain a homogeneous rate of air flow, the outlet of the ventilation system was
connected to a rectangular hood. The hood was connected to a blower with a
maximum capacity of 14 L/s through a pipe 30 mm in diameter. The flow rate of
the air was adjusted by a valve located at the site of the connection.
The number concentration and size distribution of the particles produced during the
burning process was measured using a Scanning Mobility Particle Sizer (SMPS). The
SMPS consisted of a 3071 TSI electrostatic classifier and a TSI 3010 Condensation
Particle Counter (CPC). The SMPS was operated in a window of 10 – 600 nm. The
sheath air flow was selected as 4 lpm and the sample flow was 0.4 lpm. The scanning
time and retrace time were set at 120 s and 60 s respectively. Continuous
measurements of the total particle number concentration were carried out using a
CPC TSI 3022, with the sampling interval of 20 s.
The smoke samples were taken from the flue through a probe of 1 cm in diameter
and then introduced to an ejector dilutor (Dekati) where they were diluted 10 times
199
with heated, compressed, particle-free air to obtain a dry, diluted sample and to
prevent further coagulation. The sample flow was then mixed in a dilution tunnel
with a constant flow of ambient air filtered by a HEPA filter to reduce the
concentration below 106 particles per cm3. The flow rate and temperature of the
samples in the dilution tunnel were measured using an air velocity meter. The
velocity of air in the dilution tunnel was kept higher than 1m/s in order to obtain
turbulent conditions and a good mixing of the sample. The tunnel temperature was
observed to be between 28°C and 30°C. The dilution ratio was calculated as follows:
bd
bf
CCCC
DR−
−= (5.1)
where Cf , Cd, and Cb are CO2 concentrations measured in the flue, dilution tunnel and
background, respectively (Wardoyo et al., 2006) using a flue gas analyzer and a TSI
8554 Q Trak Plus. Both the flue gas analyzer and the Q-trak were calibrated prior to
obtaining the measurements. The dilution ratio was found to vary from 100 to 200
depending on the burning conditions.
5.2.1.2 Sample Material and Preparation
The samples consisted of different species of grasses and litter collected from the
savannah of the Jabiru area in the Northern Territory, Australia, in August 2005
(which is the middle of the dry season). Three species of grasses were selected as
samples, according to their prevalence in the area (Wilson et al., 1990; Williams et
al., 1999), as well as litter, containing a mix of grasses, leaves and branches. The
grass species sampled were Shorgum intrans, Aristida holothera and Eulalia
mackinlayi. The samples were placed in plastic bags and transported from Jabiru to
ILAQH, QUT in Brisbane. Before the experiments were carried out, the moisture
200
content of the samples from each species were measured using the difference
between the sample weight before and after drying in an oven at a temperature of 110
°C. The moisture content of the samples varied from 6 – 9 % for Aristida holothera,
7 - 10 % for Eulalia mackinlayi, 6 - 11 % for Shorgum intrans, and 8 - 15 % for
litter. These moisture contents were similar to those reported for the grasses growing
in the savannas of Northern Australia during the early and the late dry season, of 19
% and 11 %, respectively (Rossiter et al., 2003). The samples were weighed for each
burning, in baches of 300 g for the grasses and 500 g for litter. The ash and the
unburned fraction of the burnt samples were weighed after the end burning.
5.2.1.3 Burning Conditions
The samples were burned in the stove under two different conditions of burning, ‘fast
burning’ and ‘slow burning’. During fast burning, the stove was connected to a
blower that introduced fresh air at a rate of 14 L/s, with the valve fully open to let air
into the stove with maximum velocity. Under slow burning conditions, the stove was
not connected to the blower and the air supply through the ventilation system was not
forced during the burning process. Burning of the samples was repeated in sequence
three times for each of the two burning conditions.
5.2.2 Airborne Measurements
The airborne measurements were carried out during two campaigns in June and
September 2003, which were in the early (EDS) and the late dry season (LDS).
201
5.2.2.1 Study area
Northern Territory
Savannas
Figure 5.1. Savannas in the Northern Territory Australia with a variety of vegetation. The black line indicates the flight path flown at various altitudes during the campaigns.
The Northern Territory, Australia, is a large tropical savannah region and has
observable wet-dry seasons. The dry season, which is mild to warm, occurs from
May to October, and this is when uncontrolled fires occur annually (Gill et al., 2000),
with fires of a mild intensity in the EDS and of a high intensity in the LDS (Williams
et al., 1998) with peak occurrence during July and September (Russell-Smith et al.,
1997).
The hot and humid wet season occurs from November to April, with an average
annual precipitation of over 1000 mm, although the annual pattern of rainfall varies
greatly from year to year, as a result of the Southern oscillation (McKeon et al.,
1990).
202
5.2.2.2 Measurement time and location
The airborne measurements of the size distribution of biomass burning particles in
the Jabiru area were conducted during campaigns in June (EDS) and September
(LDS) 2003. The measurements were performed during four flights either in the
morning or in the afternoon, with the duration of the flights from 20 to 30 minutes.
The total time spent for each measurement of the four flights was approximately four
hours, including transit from Darwin, a vertical stack of horizontal flight legs and
return transit to Darwin. In June, the conditions were mostly fine and clear while in
September, mostly fine, with occasional cloud cover.
The preference was for afternoon flights, due to the increased number and intensity
of fires during the day however this was not always possible due to cloud conditions
which essentially determined what altitudes were flown. The altitudes selected were
based on the objectives of the flight, which were either to obtain boundary layer data
alone, or a combination of boundary layer and free troposphere data, which is why
the flight leg altitudes were different between flights. The minimum altitude was set
at 0.5 km, primarily for aircraft safety reasons and the maximum altitude flown was
6.5 km.
The airborne measurements were carried out along a determined horizontal path from
a point South West (SW) of ‘Jabiru’ (13.08 South (S) 132.32 East (E)) to a point
North East of ‘Jabiru’ (12.11 oS 133.15 oE), in Kakadu National Park, Northern
Territory, Australia. The orientation of the flight path was chosen so that it was
perpendicular to prevailing wind directions on the ground. The wind direction was
203
predominantly from the South East (SE) in the morning and varied from morning to
afternoon, with the change in wind direction only influencing the original choice of
flight path orientation slightly. The majority of the fires were located on or near the
flight path, with satellite data showing 39, 28, 72 and 41 hotspots detected on the
23rd, 24th, 26th, and 27th of June 2003, respectively, and 3, 11, 11, and 6 hotspots on
the 22nd, 23rd, 25th, and 26th of September 2003, respectively
(http://www.sentinel.csiro.com.au). Although the number of fires in September was
lower, their intensity was significantly higher than those in June.
5.2.3 Data Analysis
For the laboratory experiment, the count median diameters (CMD) of particles
measured during flaming and smoldering phases for each burning of the samples
were statistically analyzed to obtain the average and standard deviation. The final
CMD and standard deviation the experiments were presented by averaging those
values.
For the airborne experiment, the average size distribution for each height region (the
definition of the regions provided below) was obtained by averaging the size
distributions measured at several heights of the region for each measurement day.
The average CMD and standard deviation for each region were then calculated from
the measured CMD for different height levels included in the area.
204
5.3. RESULTS 5.3.1 Laboratory Measurements 5.3.1.1 Particle Size Distributions
Particle size distributions during both flaming and smoldering phases of fast and
slow burning of the samples were found to unimodal, which is demonstrated in
Figure 2 presenting as an example the average particle size distribution for the slow
burning of each sample.
Figure 3 presents the average CMD and its standard deviation for particles emitted
during the flaming and smoldering phases of both fast and slow burning. For fast
burning, the CMD of particles released during the flaming phase was 55 ± 7 nm for
Aristida, 36 ± 8 nm of Eulalia, 50 ± 14 nm for Intrans and 37 ± 5 nm for litter.
During the smoldering phase of fast burning, Aristida, Eulalia, Intrans and litter
produced particles with the CMD of 46 ± 3 nm, 50 ± 4 nm, 37 ± 3 nm and 77 ± 3 nm,
respectively. Slow burning of the samples produced particles with a CMD of 122 ±
33 nm for Aristida, 161 ± 20 nm for Eulalia, 165 ± 50 nm for Intrans and 211 ± 44
nm for litter, during the flaming phase. For the smoldering phase, the CMD of
Aristida, Eulalia, Intrans and litter was 60 ± 25 nm, 87 ± 13 nm, 103 ± 25 nm and
158 ± 29 nm, respectively.
205
Arisitda
0.0E+00
5.0E+06
1.0E+07
1.5E+07
2.0E+07
2.5E+07
3.0E+07
3.5E+07
4.0E+07
1 10 100 1000
Diameter (nm)
Eulalia
0.0E+00
2.0E+07
4.0E+07
6.0E+07
8.0E+07
1.0E+08
1.2E+08
1.4E+08
1.6E+08
1.8E+08
2.0E+08
1 10 100 1000
Diameter (nm)
Parti
cle Num
ber C
once
ntra
tion
(par
ticles/cm
3 )
Flaming FlamingSmouldering Smouldering
atio
n
Figure 5.2. The average of size distribution for slow burning of grass samples.
0
50
100
150
200
250
300
Aristida Intrans Eulalia Litter Aristida Intrans Eulalia Litter
Type of Sample
CM
D (n
m)
Flaming
Smouldering
Fast Burning Slow Burning
Figure 5.3. Count median diameter (CMD) characteristic of samples burned.
It can be seen from inspecting Figure 5.3 that the CMD’s emitted during fast burning
show two distinct trends. Aristida and Intrans, emitted larger particles during the
Parti
cle Num
ber C
once
ntr
(par
ticles/cm
3 )
Litter
0.0E+00
5.0E+06
1.0E+07
1.5E+07
2.0E+07
2.5E+07
3.0E+07
3.5E+07
4.0E+07
1 10 100 1000Diameter (nm)
Parti
cle Num
ber C
oncn
etra
tion
(par
ticles/cm
3)
Intrans
0.0E+00
5.0E+07
1.0E+08
1.5E+08
2.0E+08
2.5E+08
3.0E+08
1 10 100 1000
Diameter (nm)
(par
ticles/cm
3)
FlamingFlamingSmoulderingSmouldering
atio
n Pa
rticle Num
ber C
once
ntr
206
flaming and smaller particles during the smoldering phase, while Eulalia and litter,
the other way round.
5.3.2 Airborne Measurements 5.3.2.1 Boundary Layer Measurements The boundary layer height was estimated from measurements of the temperature
profiles versus altitude and was found when the air was stable, indicated by a
temperature inversion, that is, an increase in air temperature with an increase in
height. In stable air, pollutants are trapped, which prevents them from dispersing into
the free troposphere. Figure 5.4 is an example of the temperature vertical profile
measured on 26th of June. The boundary layer was approximately 1700, 2000, 1800
and 1700 m on the 23rd, 24th, 26th and 27th of June 2003, respectively; 1800 on 22nd
of September, and 2000 m on 23rd, 25th and 26th of September 2003.
26th June
0
1000
2000
3000
4000
5000
0 5 10 15 20 25 30 35 40
Temperature (o C)
Alti
tude
(m)
Figure 5.4. The temperature vertical profile measured on the 26th of June 2003.
207
5.3.2.2 Particle Size Distributions In order to characterize the particle size distribution obtained during the campaigns,
the measurements were classified into three height level regions, being region I - the
lower boundary layer (LB, calculated to be about 1700 m for the June campaign and
1900 m for the September campaign),), region II - the upper boundary layer (UB,
heights between 2000 and 3900 m), and region III - free troposphere (FT, height
greater than 3900 m).
Figure 5. Average size distributions for June and September campaigns
Region I
0
500
1000
1500
2000
2500
1 10 100 1000Diameter (nm)
dN/dlogD
p (p
artic
les/cm
3 )
June Campaign
September Campaign
Region II
0
200
400
600
800
1000
1200
1 10 100 1000Diameter (nm)
dN/dlogD
p (p
artic
les/cm
3 )
June CampaignSeptember Campaign
Region III
0
200
400
600
800
1000
1200
Figure 5.5. Average size distribution for the June and September campaigns.
Figure 5.5 shows the average of the size distribution of particles measured during the
June and September campaigns for the three regions. For the June campaign (EDS),
the peak of particle size distribution is broad across the Region I and II, with a
1 1 100 1D
0 000iameter (nm)
dN/dlogD
p (p
artic
les/cm
3 )
June CampaignSeptember Campaign
208
significantly larger concentration of particles in the lower level (Region I). The size
distribution of particles was significantly influenced by local plumes of fresh smoke
from nearby fires, as shown by satellite data which indicated many of the fires
located close to the flight path (http://www.sentinel.csiro.com.au). During the
September campaign (LDS), the peak of the size distribution is narrow, indicating
that the particles had more homogeneous diameters across all three regions. This
implies that the particles measured during the LDS campaign were present in the
atmosphere for some time prior to the measurements, which enabled air mixing and
particle dynamics processes to occur. The hotspot satellite data showed that high
intensity fires were detected of the order of hundreds of kilometers away from the
flight path (http://www.sentinel.csiro.com.au), thus the particles had likely originated
from fires that occurred before the day of the measurement.
0
20
40
60
80
100
120
140
160
0 1000 2000 3000 4000 5000 6000 7000
Height (m)
CM
D (n
m)
June Campaign
September Campaign
Region I Region II Region III
Figure 5.6. The measured CMD of particles during June and September campaigns.
209
Figure 5.6 shows the average and standard deviation of CMDs of particles measured
during the June and September campaigns, versus height of the measurements. In
general, it can be seen that CMD’s are larger at lower levels and decrease with
height. For the June campaign the CMDs of particles in region I was relatively large,
with a large standard deviation due to the proximity to the source, which means a
number of fires occurring close to the flight paths. The CMD decreased in region II
and then further decreased in region III. During the September campaign, the CMDs
for regions I and II were larger than those for the June campaign and similar for both
regions, but decreased significantly in region III, to the similar values as recorded in
June. The standard deviation of CMD was relatively small, pointing out again to well
mixed, aged aerosol.
For the June campaign, the average and standard deviation of the CMD in region I,
II, and III was 83 ± 13 nm, 68 ± 10 nm, and 56 ± 2 nm, respectively. For the
September campaign, the CMD was found to be 127 ± 6 nm in region I, 119 ± 10 nm
in region II, and 59 ± 7 nm in region III. From these figures, it was found that the
CMD decreased from region I to region II, and also from region II to region III by
18%, for the June campaign. For the September campaign, the CMD decreased by
6% from region I to region II and a significantly larger decrease in CMD of 50 %
was found from region II to region III.
5.4. DISCUSSION AND CONCLUSIONS
5.4.1 Particle Diameter
This study presents the results of investigations of particle size distribution
originating from burning of several grass species under controlled laboratory
210
conditions, and also at several heights during field campaigns conducted during dry
seasons in the Northern Territory, Australia. The laboratory study simulated real field
conditions, such as burning phases and burning rate, as accurately as possible. Under
real field conditions, biomass burning occurs mainly in the flaming and smoldering
phases, with most emissions produced during these phases, and therefore laboratory
measurements were conducted during these phases. Biomass burning also occurs
under a variety of wind speeds, which is associated with variation in burning rate. In
order to simulate this, the laboratory study was also set up to include different
burning rates, namely fast and slow burning.
5.4.1.1 Laboratory Studies
The results from the laboratory study showed that burning rate has an important
impact on the characteristics of the particle emissions. Fast burning released smaller
particles during burning process, with an average CMD of 50 nm, across the different
grass species. Whilst slow burning results in larger particles, with CMD of 165 and
100 nm during the flaming and smoldering phases, respectively. This can be
explained by the introduction of fresh air into the burning system, during fast burning
whereby the increased oxygen supply involved in the burning process causes a more
complete burning and in turn production of smaller particles. On the other hand, a
lesser supply of oxygen into the burning system during slow burning caused the
burning process to be incomplete and consequently more smoke and release of larger
particles. This phenomena has also been reported by a recent study investigating size
distribution of particles released during burning of several types of trees growing in
Queensland, Australia (Wardoyo et al., 2006). The study reported that during fast
211
burning, the diameter of particles emitted by wood burning were 40 nm for flaming
and 35 nm for smoldering, whilst the diameter of particles emitted by leaf and branch
burning were 50 nm for flaming and 35 nm for smoldering. During slow burning, the
diameter of particles emitted for wood burning were 50 nm for flaming and 35 nm
for smoldering, and for leaf and branch burning, the diameter of particles emitted
during flaming and smoldering were 150 nm and 50 nm, respectively. By
comparison to the previous study conducted by Wardoyo et al, 2006, this study found
larger particles during flaming and smoldering for fast burning, with the exception of
the flaming phase for burning leaf and branch. For slow burning, this study also
found larger particles than those measured in the previous study. Based on the results
from both studies, the diameter of particles was found to be significantly dependant
on the type of vegetation.
A study conducted by Hueglin and colleagues (1997), using a residential wood stove
to measure size distribution of particles emitted during burning also demonstrated
that particles emitted are of different sizes, according to the different phases of
burning. Burning of beech wood, with moisture content between 15 – 18 % resulted
in particles with diameters around 170 and 60 nm during the flaming and smoldering
phases, respectively. The study used an automatically operated wood chip burner to
obtain varying burning conditions and also found that the air supply affected the size
distribution of particle emissions, whereby an increased air (oxygen) supply
produced smaller particles, relative to those produced when air (oxygen) supply was
limited (Hueglin et al., 1997).
212
A similar study, investigating particles emitted during burning of birch wood with
the moisture content of 15- 18 %, using a commercial soapstone stove operated with
the burn rate of 3.5 kg of fuel/h, reported that most of the emitted particles had a
diameter in the range from 30 to 130 nm, and observed that the particle size
distribution and number concentration depended on the phase of the burning process
and the burning rate (Hedberg et al., 2002).
A study conducted by Wieser and Gaegauf (2005) using several combustion systems
with varying burning rates showed that the variation in particle size distribution
depended on the type of combustion system. This study also investigated the effect
of excess air on particle formation, and reported that less air (oxygen) supply resulted
in large particles with small number were released (Wieser and Gaegauf, 2005).
Hays and colleagues (2002) studied the characteristics of the particle size distribution
by simulating open burning in an enclosure (~ 28 m3), with an interior wall lined
with aluminum foil and with constant air circulation (~ 20 m3 per min). Sand-lined
(~ 2.5 cm) stainless steel pan (0.8 m2) was used as the firebox, and positioned on an
electronic platform balance to monitor the burning mass of the fuels. Fresh green
foliage and litter, of species collected from native US habitats, frequently
experiencing fire, were used as fuels. The study reported that the size distribution
was dependant on the phase of the burning process, with particle diameters of 100 to
150 nm for the flaming phase and 70 to 150 nm for the smoldering phase, depending
on the fuel (Hays et al., 2002).
In summary, the reported laboratory studies found significant variation in particle
characteristic dependently of the burning conditions, with smaller particles produced
213
during fast burning and large during slow burning. There were no major differences
in particle diameter reported for different burning phases of fast burning process,
however for slow burning, the size distribution of particles depended on the burning
phases, with larger particles emitted during flaming and smaller during smoldering
phases, respectively. The results of this study confirmed these general trends and
quantified particle size characteristics resulting from burning of a variety of grass
species growing in the Northern Territory, Australia.
5.4.1.2 Field Studies
The study found that the characteristics of particles emitted during different parts of
the dry season were different, with smaller particles measured in the EDS and larger
in the LDS. The study also investigated a vertical profile of particles, and
demonstrated similar trends for both EDS and LDS of the diameter decreasing with
the increasing height.
A number of studies measuring particle size distribution from biomass burning have
been conducted around the world, with previous studies showing that particles
measured over African savannah had a CMD of 220 ± 3 nm (Anderson et al., 1996),
over South Africa of 120 ± 250 nm (Le Canut et al., 1996), over temperate forests of
North America, of 190 ± 3 nm (Hobbs et al., 1996), and of aged smokes in Brazil
ranging from 120 to 230 nm (Anderson et al., 1996; Reid et al., 1998). Two separate
studies conducted in Amazonia reported different ranges of particle diameter: 15 to
279 nm (Guyon et al., 2005) and 51 to 144 nm (Krejci et al., 2005).
214
In summary, it can be concluded that particle size distribution varies between the
different locations due to a variety of factors, including the differences in vegetation,
its moisture contents, and weather conditions during burning. The CMDs of particles
measured during field campaigns in the Northern Territory, Australia, conducted as a
part of this study were comparable to the lower ranges of those reported in the
previous studies, ranging from 56 to 127 nm across both EDS and LDS.
5.4.1.3 Comparison between CMD measured in the laboratory and in the field
Count median diameter (CMD) of particles measured during field campaigns
conducted over Northern Territory in region I (which is at the lowest heights, which
means the closest to where the particles were generated) was 83 nm and 127 nm
during the EDS and the LDS, respectively. The laboratory measurements found the
diameter of particles resulting from the grass species growing in the Northern
Territory varied in CMD from 50 nm to 165 nm, depending on the rate of burning.
These two ranges are quite comparable, despite the unavoidable differences between
laboratory and field conditions. In particular, laboratory measurements were carried
out with simulated wind speeds of 2 m/s for slow burning and 20 m/s for fast
burning, while for most times during the field measurements wind speed was of 5
m/s. Since the increase in wind speed, from 2 to 20 m/s, resulted in an increase in
particle CMD from 50 nm to 165 nm, and assuming a linear relationship between
these two parameters, it could be expected that for a wind speed of 5 m/s, the particle
CMD would be approximately 70 nm. This estimated diameter is comparable to the
CMD of particles measured in region I, during the early dry season. Of course, the
estimation can not be used as a verification of the accuracy of the results, because
215
there are many different factors influencing the conditions in the laboratory and the
field.
5.4.2 Particle Vertical Profile
The airborne measurements of particle size distribution during the two field
campaigns also provided valuable information on the vertical profile of particles in
the atmosphere. The vertical profile of particles in EDS and LDS has been discussed
above, in terms of vertical size distribution, however, the vertical profile of particles
also can be viewed in terms of the change in particle concentration. Using the data on
particle concentration measured during both campaigns, it was found that the
concentration of particles decreased by 53 % and 52 % in region II in comparison
with region I, for the June and September campaigns, respectively (Ristovski et al.,
2006). The decrease in particle concentration between these two regions is mostly
influenced by the presence of the boundary layer and only limited penetration of
particles through the boundary layer. Therefore, it can be concluded that there is no
significant difference in the relative change in particle concentrations between these
two regions. However, the reduction in CMD between these two regions, that is 20%
in the EDS and 6% in the LDS, clearly shows a difference in particle characteristics
with height. In the EDS, the particles measured were smaller and CMD reduced with
the increased height. The reduction in CMD from below (region I) to above (region
II) the boundary layer indicated that the particles in each region have different
physical properties. On the other hand, there was no significant difference in CMD
below or above the boundary layer in the LDS, indicating similar physical properties
of particles in both regions. This implies that the particles measured during the LDS
216
were suspended in the atmosphere for long enough, enabling efficient mixing and
achieving uniform condition in terms of particle characteristics below and above the
boundary layer.
Comparison of particle characteristics between regions II and III showed that the
there was no significant differences in particle concentration during the EDS, and a
small but statistically significant decrease in concentration of 13% in the LDS
(Ristovski et al., 2006). However, there were larger differences in particle CMD,
with the decrease by 18 % in the EDS and by 51 % in the LDS between regions II
and III. For the EDS, the characteristics of the particles found in region II and III
indicate that they were mostly existing particles, called ‘free troposphere particles’.
For the LDS, a large difference in particle CMD between the region II and III was
found, indicating that the particles in these two regions have different physical
properties. The large particles found in region III during the LDS indicated that those
particles were a mixture between existing particles (free troposphere particles) and
particles from other sources.
5.4.3. Particle Age The change of the particle size distribution during the transport in the air is affected
by simultaneous generation and coagulation of particles and smoke mixing with the
ambient air (Snegirev et al., 2001). The distribution is given by the dynamic equation
(Seinfeld, 1986)
( ) ( ) ( ) ( ) ( ) ( ) vdtvnvvtvnvdtvntvvnvvvt
tvn v
′′′Γ−′′′−′−′Γ=∂
∂∫∫∞
,,,,,,21),(
00
( ) ( tvntvn
d
,,0&+−
τ) , (5.2)
217
Where is the particle size distribution, ),( tvn ( )vvv ′−′Γ , is the coagulation
coefficient, dτ is the characteristic time scale of particle dilution by ambient air, is
the particle generation rate, and t is the residence time which is the time of
movement from the generation zone to the detection point or the age of particles. The
dilution time and the age of particles depend on the condition of turbulent mixing,
flow intensity, and geometry of compartment (Snegirev et al., 2001). Provided that
the coagulation coefficient is constant, the differential equation of the particle size
distribution can be derived from equation (5.2)
0n&
0
2
2dtd NNNN
d
&+−Γ
−=τ
, (5.3)
Where N is the total particle concentration, is the total particle generation rate.
Assuming the rate of the particle dilution and the particle generation are equal, and
coagulation is the only factor influencing the characteristic of the particle size
distribution during the transport, the total particle concentration can be presented by
the coagulation equation (Hinds, 1982):
0N&
K tN 1N
(t) No
o
+= (5.4)
where N(t) is the total particle concentration, K is the coagulation coefficient, No is
the initial particle concentration.
There are many other factors that could play a role in changing of the particle size
distribution in field, such as: dispersion of plume, sedimentation of particles, and
variations in fire size and intensity (Radke et al., 1995). However to simplify and in
fact enable this assessment, those factors are neglected. By using the data of the
particle concentrations measured during the campaigns presented in Table 5.1, the
218
age of particles and the initial particle concentration are calculated using the equation
(5.4) with the coagulation coefficient corrected as function of the diameter (Hinds,
1982). The age of particles and the initial particle concentration are found 1.2 ± 0.5
days and 975 ± 40 particles/cm3 for the EDS, and 6.5 ± 0.5 days and 2995 ± 260
particles/cm3 for the LDS respectively. The result shows that particles found during
the EDS were approximately 5 days younger than those measured in the LDS. This
can be explained by the fact that those particles originated from fresh smoke coming
from the sources that were close to the flight paths during the measurements. The age
of biomass burning particles reported in other studies was found to be about 2 days
(Radke et al., 1995) and 5 – 15 days (Reid et al., 1999a).
Table 5.1. Particle concentration measured during the campaigns
Campaign Day of
Measurement Particle concentration in
region I (#/cm3) 23-Jun-03 898 ± 298 24-Jun-03 913 ± 1000 26-Jun-03 581 ± 179
JUNE
27-Jun-03 506 ± 104 22-Sep-03 1393 ± 231 23-Sep-03 1255± 309 25-Sep-03 823 ± 91
SEPTEMBER
26-Sep-03 1280 ± 180
219
5. 5. REFERENCES
Abel, S., J. M. Haywood, J. Li and P. R. Buseck (2003). Evolution of biomass
burning aerosol properties from an agricultural fire in south Africa.
Geophysical Research Letters, 30(15): 1783, doi:10.1029/2003GL017342.
Anderson, B. E., W. B. Grant, G. L. Gregory, E. V. Browell, J. E. Collins Jr, D. W.
Sachse, D. R. Bagwell, C. Hudgins, D. R. Blake and N. J. Blake (1996).
Aerosols from biomass burning over the tropical South Atlantic region:
Distributions and impacts. Journal of Geophysical Research, 101: 24117-
24137.
Bodhaine, B. A. (1983). Aerosol measurements at four background sites. Journal of
Geophysical Research, 88: 10753 - 10768.
Cofer III, W. R., E. L. Winstead, B. J. Stock, L. W. Overby, J. G. Goldammer, D. R.
Cahoon and J. S. Levine (1996). Emissions from boreal forest fires: are the
atmospheric impacts underestimated ?, in: Global Biomass Burning and
Global Change, edited by Levine, J.S., MIT Press, Cambridge, Mass, pp 716-
7332.
Ferge, T., J. Maguhn, K. Hafner, F. Muhlberger, M. Davidovic, R. Warnecke and R.
Zimmermann (2005). On-line analysis of gas phase composition in the
combustion chamber and particle characteristics during combustion of wood
and waste in a small batch reactor. Environmental Science and Technology,
39(6): 1393-1402.
Gao, S., D. A. Hegg, P. V. Hobbs, T. W. Kirchstetter, B. I. Magi and M. Sadilek
(2003). Water soluble organic components in aerosols associated with
savanna fires in southern Africa: identification, evolution, and distribution.,
220
Journal of Geophysical Research, 108(D13): 8491, doi:
10.1029/2002JD002324.
Gill, A. M., P. G. Ryan, P. H. R. Moore and M. Gibson (2000). Fire regimes of world
heritage Kakadu National Park Australia. Australia Ecology, 25: 616-625.
Glasmann, I. (1988). Soot formation in combustion processes. In Twenty-Second
Symposium (International) on combustion, The combustion Institute,
Pittsburgh, p.295.
Gras, J. L. (1991). Southern hemisphere tropospheric aerosol microphysics., Journal
of Geophysical Research, 96(D3): 5345-5365.
Gras, J. L. (1999). Some optical properties of smoke aerosol in Indonesia and
Tropical Australia. Geophysical Research Letters, 26(10): 1393-1396.
Guyon, P., G. Frank, M. Welling, D. Chand, P. Artaxo, L. Rizzo, G. Nishioka, O.
Kolle, H. Fritsch, M. A. F. S. Dias, M. Cordova and M. O. Andreae (2005).
Airborne measurements of trace gas and aerosol particle emissions from
biomass burning in Amazonia. Atmospheric Chemistry and Physics, 5: 2791-
2831.
Hays, M. D., C. D. Geron, K. J. Linna, N. D. Smith and J. J. Schauer (2002).
Speciation of gas-phase and fine particle emissions from burning of foliar
fuels. Environmental Science and Technology, 36: 2281-2294.
Haywood, J. M., S. R. Osborne, P. N. Francis, A. Keil, P. Formenti, M. O. Andreae
and P. H. Haye (2003). The mean physical and optical properties of regional
haze dominated by biomass burning aerosol measured from the C-130 aircraft
during SAFARI 2000., Journal of Geophysical Research, 108(D13): 8473.
221
Hedberg, E., A. Kristensson, M. Ohlsson, C. Johansson, P.-A. Johansson, E.
Swietlicki, V. Vesely, U. Wideqvist and R. Westerholm (2002). Chemical
and physical characterization of emissions from birch wood combustion in a
wood stove. Atmospheric Environment, 36(30): 4823-4837.
Hinds, W. C. (1982). Aerosol Technology: Properties, Behavior, and Measurement
of Airborne Particles, pp:263-265, A Wiely-Interscience Publication, LA,
California.
Hobbs, P. V., J. S. Reid, J. A. Herring, J. D. Nance, R. E. Weiss, J. L. Ross, D. A.
Hegg, R. D. Ottmar and C. Liousse (1996). Particle and trace gas
measurements in smoke from prescribed burns of forest products in the
Pacific Northwest , in: Biomass Burning and Global Change, Vol 1,edited by
:Levine, J.S, MIT Pres, New York.: p 618-636.
Hueglin, C. H., C. H. Gaegauf, S. Kunzel and H. Burtscher (1997). Characterization
of wood combustion particles: morphology, mobility, and photoelectric
activity. Environmental Science and Technology, 31: 3439-3447.
Krejci, R., J. Strom, M. Reus, de, J. Williams, H. Fischer, M. O. Andreae and H. C.
Hansson (2005). Spatial and temporal distribution of atmospheric aerosols in
the lower troposphere over the Amazonian tropical rain forest., Atmospheric
Chemistry and Physics, 5: 1527-1543.
Le Canut, P., M. O. Andreae, G. M. Harris, F. G. Weinhold and T. Zenker (1996).
Airborne studies of emissions from savanna fire in southern Africa, 1,
Aerosol emissions measured with a laser optical particle counter. Journal of
Geophysical Research, 101: 23615-23630.
222
McKeon, G. M., K. A. Day, S. M. Howden, J. J. Mott, W. J. Orr, W. J. Scattini and
E. J. Weston (1990). Northern Australia Savannas: management for pastoral
production. Journal of Biogeography, 17: 355-372.
Nichol, J. (1997). Bioclimatic impacts of the 1994 smoke haze event in southeast
Asia. Atmospheric Environment, 31(8): 1209-1219.
Radke, L. F., D. A. Hegg, P. V. Hobbs and J. E. Penner (1995). Effects of aging on
the smoke from a large forest fire. Atmospheric Research, 38: 315-332.
Reid, J. S., T. F. Eck, S. A. Christopher, P. V. Hobbs and B. R. Holben (1999a). Use
of the Angstrom exponent to estimate the variability of optical and physical
properties of aging smoke particles in Brazil., Journal of Geophysical
Research, 104: 27489-27498.
Reid, J. S. and P. V. Hobbs (1998). Physical and optical properties of smoke from
individual biomass fires in Brazil. J. Geophys. Res, 103: 32013-32031.
Reid, J. S., P. V. Hobbs, A. L. Rangno and D. A. Hegg (1998). Relationships
between cloud droplet effective radius, liquid water content, and droplet
concentration for warm clouds in Brazil embedded in biomass smoke. Journal
of Geophysical Research, 104: 6145-6153.
Reid, J. S., P. V. Hobbs, A. L. Rangno and D. A. Hegg (1999b). Relationships
between cloud droplet effective radius, liquid water content, and droplet
concentration for warm clouds in Brazil embedded in biomass smoke. Journal
of Geophysical Research, 106: 6145-6153.
Ristovski, Z., A. Y. P. Wardoyo, L. Morawska, M. Jamriska, S. Carr and G. Johson
(2006). Biomass burning influenced particle characteristic in Northern
Territory of Australia based on Airborne Measurements.
223
Rossiter, N. A., S. A. Setterfield and M. M. Douglas (2003). Testing the grass fire
cycle:alien grass invasion in the tropical savannas of northern Australia.
Diversity and Distributions, 9: 169-176.
Russell-Smith, J., P. B. Ryan and R. Durieu (1997). A LANDSAT MSS-derived fire
history of Kakadu National park, monsoonal Australia, 1980-94:seasonal
effect, frequency and patchiness. Journal of Applied Ecology, 34279-87.
Seinfeld, J. H. (1986). Atmospheric chemistry and physics of air pollution., New
Yorks: Wiley.
Shaw, R. W. (1987). Air pollution by particles. Science Environment, 255: 96 - 103.
Snegirev, A. Y., G. M. Makhviladze and J. P. Roberts (2001). The effect of particle
coagulation and fractal structure on the optical properties and detection of
smoke. Fire Safety Journal, 36(1): 73-95.
Tsutsumi, Y. (1999). Aircraft measurement of ozone, NOx, CO, and aerosol
concentrations in biomass burning smoke over Indonesia and Australia in
October 1997: Depleted ozone layer at low altitude over Indonesia.,
Geophysical Research Letters, 26(5): 595-598.
Wardoyo, A. Y. P., L. Morawska, Z. Ristovski and J. Marsh (2006). Characterization
particle number and mass emission factors from combustion Queensland
trees., Environmental Science and Technology, 40: 5696-5703.
Wieser, U. and C. k. Gaegauf (2005). Nanoparticle emissions of wood combustion
processes, Laboratories for Sustainable Energy System, Accessed March
2005,
http://www.oekozentrum.ch/downloads/publikationen/nanoparticles.pdf.
224
Williams, R. J., G. D. Cook, A. M. Gill and P. H. R. Moore (1999). Fire regime, fire
intensity and tree survival in a tropical savanna in northern Australia.
Australian Journal of Ecology, 24: 50-59.
Williams, R. J., A. M. Gill and P. H. R. Moore (1998). Seasonal changes in fire
behavior in a tropical savanna in northern Australia. International Journal of
Wildland Fire, 8: 227-239.
Wilson, B. A., P. S. Brocklehurst, M. J. Clark and K. J. M. Dickinson (1990).
Vegetation survey of the Northern Territory Australia. Technical Report no:
49, Conservation Commission of the Northern Territory, Darwin.
Wurzler, S. and M. Simmel (2005). Impact of vegetation fires on composition and
circulation of the atmosphere, Accessed March 2005,
http://projects.tropos.de:8088/afo200g3/.
225
226
CHAPTER 6. GENERAL DISCUSSION
6.1. INTRODUCTION
As discussed in Chapter 1, biomass burning affects large areas worldwide. Millions
of hectares are destroyed by biomass burning every year, including controlled and
uncontrolled burning, savannah fires, wildfires, and burning of prescribed wild land,
agricultural, logging slash and land clearing slash. Savannas and tropical and sub
tropical forests are mostly burned annually, with millions hectares of areas
consumed.
6.1.1. Biomass burning emissions Biomass burning around the world produces emissions that have been recognised as
a major contributor to atmospheric particulate matter and gases. Biomass burning
contributes to 38 % of the particulate matter, 40 % of the CO2, 32 % of the CO, 24 %
of the NOx, 21 % of the NH3, and 25 % of the ozone released by all sources
(Andreae, 1991). Other data show the contribution of biomass burning to the global
atmospheric budgets of CO is 59 % of the total from all sources with an emission
rate of 748 Tg/year (Uherek, 2004). Biomass burning in Texas in 1996 emitted
161,000 tons/year of particulate matter and 698,000 tons/year of gases including CO,
CH4, NOx, and NH3 (Dennis et al., 2002). Wood burning in Sweden produces 8,600
to 65,000 tons/year of particulate matter every year (Areskoug et al., 2000). Burning
of cereal wastes in Spain releases total particulate matter (TPM) of 80 – 130 Gg, NOx
of 17 – 28 Gg, CO of 210 – 350 Gg, and CO2 of 8 – 14 Gg annually (Ortiz de Zarate
227
et al., 2000). Significantly large amounts of emissions from biomass burning
released to the atmosphere show that biomass burning has become a serious problem,
associated with serious effects.
6.1.2. Biomass burning particles Knowledge of the characteristics of biomass burning particle emissions in terms of
size distribution and emission factors, particularly in terms of particle number
emissions, has been identified as a very important element in developing a
quantitative assessment of its impacts. Biomass burning particles were reported in a
variety of size distributions. The majority of particles were less than 2.5 μm in
diameter or fine particles, PM2.5 (Hueglin et al., 1997; Hays et al., 2002; Hedberg et
al., 2002; Reid et al., 2005), and ultrafine particles with diameters less than 0.1 μm
(Ferge et al., 2005; Wieser and Gaegauf, 2005).
The studies reported in the literature show that PM2.5 emission factors vary
depending on the species of tree. A study of an open burning of mixed hardwood
forest foliage in the US showed that the PM2.5 emission factors were 10.8 ± 3.9 g/kg
(Hays et al., 2002). PM2.5 emission factors from the burning of wood grown in the
North-eastern United States were measured in the range of 2.7 to 5.7 g/kg for hard
woods and 3.7 to 11.4 g/kg for soft woods (Fine et al., 2001). A similar study of the
wood grown in the Southern United States yielded emission factors in the range of
3.3 to 6.8 g/kg for hard woods and 1.6 to 3.7 g/kg for soft woods (Fine et al., 2002).
A study aimed at characterization of emissions from the burning of wood in a
fireplace found that the emission factors were 2.9 to 9 g/kg for softwoods and 2.3 to
228
8.3 g/kg for hardwoods (McDonald et al., 2000). The burning of Birch wood in a
stove produced particles with emission factors of 0.1 to 2.6 g/kg (Hedberg et al.,
2002). The emission factors from burning of wood logs in several combustion
systems were reported in the range of 0.13 to 1.68 g/kg (Wieser and Gaegauf, 2005).
However the existing data on PM2.5 emission factors are still very limited, with data
unavailable for many types of biomass and places in the world that frequently
experience fires.
For ultrafine particles, since their size is less than 0.1 μm, measurements are
conducted in terms of particle number emission factor rather then particle mass
emission factor. However, there has been only one study reporting particle emission
factors, with results of 1.43 to 39.5 ×1016 particles/kg for the burning of unspecified
woods (Wieser and Gaegauf, 2005). Knowledge of particle number emission factors
from biomass burning is essential in assessing its effects. In the case of human
health, the number of particles that penetrate and deposit within the respiratory
system may determine subsequent health effects. Particle number is also an
important factor in atmospheric processes, such as coagulation, deposition,
nucleation, and condensation. Lack of understanding of the characteristics of
biomass burning particles, high particle number concentrations in the field, and
complexity in measurement systems may be the cause of the very limited data on
particle number emission factors. Moreover there are unavailable data for most
aspects of biomass burning.
229
6.1.3. Biomass burning impacts Impacts on atmospheric process Biomass burning emissions have been known to affect atmospheric processes.
Biomass burning produces greenhouse gasses CO2 and CH4, heating the atmosphere
through absorption of thermal radiation (Levine, 1991). The emissions of CO, CH4
and volatile organic compounds (VOC) affect the oxidation capacity of the
troposphere by reacting with OH radicals, nitric oxide (NO) and VOC that lead to
formation of ozone and other photo oxidants (Koppmann et al., 2005). The emission
of methyl bromide (CH3Br) causes the photochemical destruction of ozone in the
stratosphere (Andreae, 1991). In terms of biomass burning particle emissions, their
presence in the air significantly affects atmospheric processes (Bodhaine, 1983;
Shaw, 1987). Such affects could include acidifying clouds, rain, and fog (Nichol,
1997); altering microphysical cloud processes on a small scale and mesoscale; and
altering the radiation balance of the earth, both directly, by absorbing and scattering
incoming solar radiation, and indirectly, by acting as cloud condensation nuclei
(CCN) (Kaufman et al., 1998; Martins et al., 1998; Wurzler and Simmel, 2005).
Impacts on human health Biomass burning emissions (particles and gasses) have impacts on human health.
The World Health Organization (WHO) identifies a number of emissions from
biomass burning that affect human health, and divides these emissions into classes:
particulate matter, polynuclear/polycyclic aromatic hydrocarbons (PAH), carbon
monoxide (CO), aldehydes, organic acids, semi-volatile and volatile organic
compounds, nitrogen and sulphur-based compounds, ozone and photochemical
230
oxidants, inorganic fraction of particles, and free radicals. This study focuses on
particulate matter as the topic of interest. Particulate matter has been recognized to
have serious impacts on human health and is linked to morbidity and mortality
presented in previous epidemiological and toxicological studies.
An epidemiological study conducted by the American Cancer Society in 1995
investigated the association between fine particles PM2.5 and premature death caused
by cardio-pulmonary failure. The report showed that the difference in mortality was
17 % for the difference in PM2.5 between the cleanest and dirtiest cities of 24.5 µg/m3
(Pope et al., 1995). Goldberg and colleagues used the details of recorded health data
from patients to analyze the correlation between particulate matter and mortality in
Montreal in order to link individual death to medical information up to five years
before death. The result showed that death related to cancer and chronic coronary
artery disease was associated with the concentration of PM2.5 (Goldberg et al., 2000).
Other studies also showed the relationship between particulate matter, PM2.5, and
morbidity and mortality caused by asthma (Vedal et al., 1998; Norris et al., 1999;
Tolbert et al., 2000; Yu et al., 2000), bronchitis (Etzel, 1999; McConnel et al., 1999;
Peters, J.M et al., 1999), and heart disease (Peters, A et al., 1999; Peters et al., 2000).
Ultrafine particles have been recognized to have serious impacts on human health
due to their ability to induce inflammation per unit particulate matter mass arising
from their high particle numbers, greater lung deposition rates, and surface chemistry
through reactive oxygen species (ROSs). A study found that the deposition
efficiency of ultrafine particles in human subjects was more than 60 % (Chalupa et
231
al., 2004). Ultrafine particles have been reported to be capable of inducing
pulmonary inflammation, as well as entering the cardiovascular system (Oberdorster,
2001; Nemmar et al., 2002; Oberdorster et al., 2002; Nemmar et al., 2004). In other
terms of toxicological basis, organic compounds such as polycyclic aromatic
hydrocarbons (PAHs), have been shown to induce a broad polyclonal expression of
cytokines and chemokines in respiratory epithelium. Previous studies have also
shown that ultrafine particles constitute the largest fraction of PAHs (Elguren-
Fernandez et al., 2003; Li et al., 2003). In addition, PAHs, metal, and other related
compounds may lead to the production of cytotoxic ROSs (Nel et al., 1998; Nel et
al., 2001) that induce oxidant injury and inflammatory response (Pritchard et al.,
1996). Dhalla et al. 2000, found the importance of oxidant stress responses to
cardiovascular effects (Dhalla et al., 2000). Li and colleagues showed that ultrafine
particles were most potent toward inducing cellular heme oxygenase-1 (HO-1)
expression and depleting intracellular glutathione (Li et al., 2003).
Epidemiological studies of ultrafine particles are very complicated involving several
factors that need to be considered: physical and chemical properties of ultrafine
particles, sources of ultrafine particles, behaviour of ultrafine particles in the air after
emission from sources, indoor and outdoor concentrations in the places where people
typically live related to the dose, monitor placement for ultrafine particles since most
monitoring places are far away from the particle sources, and analysis techniques for
the studies. Previous epidemiological studies have recognized the link between
exposures of ultrafine particles and health effects. The reports showed a strong
association between ultrafine particles and respiratory health in asthmatic adults
232
(Peters et al., 1997; Von Klot et al., 2000) and among children (Pekkanen et al.,
1997). Positive correlations of cardiovascular mortality with ultrafine particles were
found in an epidemiological study conducted by Wichmann and colleagues
(Wichmann et al., 2000). There have been reports that ultrafine particles have
contributed to other epidemiological evidence of adverse effects on the
cardiovascular system (Oberdorster et al., 1995; Seaton et al., 1999; Delfino et al.,
2005). A study conducted in Erfurt, Germany in 1995-98 using Poisson regression
techniques with generalized additive modelling (GAM) to analyse the data found a
positive association between ultrafine particle concentration and mortality
(Wichmann and Peters, 2000).
6.1.4. Characteristics of biomass burning particles Characteristics of biomass burning particles have been reported depending on the
type of vegetation and its moisture content. However, characteristics of biomass
burning particles are not well understood due to the complexity of influencing factors
and natural conditions in fields. The relationships between many factors and
characteristics of biomass burning particles are not yet clear. These include the
diversity of vegetation; in which every biomass consists of a variety of compounds
such as cellulose, hemicelluloses, lignin, and proteins as a result of photosynthesis
processes; and the complexity of the burning process; that involves physical and
chemical reactions, and heat and mass transfers.
Characteristics of biomass burning particles have been identified as dependant on the
burning phases of ignition, flaming, and smoldering. Previous studies reported that
233
the size of biomass burning particles was proportional to flame size (Glasmann,
1988) and fire intensity (Cofer III et al., 1996; Reid, J. S and Hobbs, P. V, 1998), and
also dependant on burning phase (Hueglin et al., 1997; Hedberg et al., 2002).
However quantitative knowledge of the relationships between these factors and
particle characteristics is still very limited. The effects of the combination of the two
in relation to characteristics of biomass burning particles still requires further
investigation. Furthermore, the effects of the amount of oxygen and rate of oxygen
supply during the burning process on characteristics of particles are poorly
understood. Consequently, the relationship between the characteristics of particles
and each phase of burning is still unclear.
Particles from biomass burning have previously been reported to vary in particle size
distribution (Hueglin et al., 1997; Hays et al., 2002; Hedberg et al., 2002; Wieser and
Gaegauf, 2005). Emitted particles have been reported to experience growth during
transport in the atmosphere and increase in size with age. The growth of biomass
burning particles begins as soon as 0.5 hours after emission, and occurs on time
scales of days (Hobbs et al., 1996; Reid et al., 1998). Biomass burning particles
found in aged smoke were generally in the presence of secondary particles
containing organic acids that enriched the process of particle growth (Andreae et al.,
1998; Formenti et al., 2003; Gao et al., 2003). However there are a number of issues
that are poorly understood: the physical and chemical characteristics of particles,
growth mechanisms, factors and processes influencing particle growth, and processes
occurring during their transport in the atmosphere.
234
Particles from aged smoke have been measured in several areas to characterize the
properties of particles in the regions, including Africa (Anderson et al., 1996; Le
Canut et al., 1996; Dubovik et al., 2002; Eck et al., 2003; Haywood et al., 2003),
North America (Dubovik et al., 2002; Eck et al., 2003), South America (Andreae et
al., 1988; Reid, J.S and Hobbs, P.V, 1998; Reid et al., 1998; Dubovik et al., 2002;
Eck et al., 2003), Europe (Fiebig et al., 2003), and the Mediterranean (Formenti et
al., 2002). The reports showed that particle size varied depending on the region
where the measurements were conducted. However the factors influencing these
differences are still unclear and speculative.
The facts show that the characteristics of particles emitted by biomass burning in
fire-prone states of Australia are poorly known due to the limited investigation
carried out in these regions. Previous measurements have been mainly over the
Eastern part of the continent (Gras, 1991). A few campaigns have focused on
characterizing the biomass burning smoke in both the Northern Territory of Australia
and in parts of Indonesia (Borneo), mainly concentrating on measurements of the
scattering coefficient, enabling differences to be highlighted between the two regions
(Gras, 1999; Tsutsumi, 1999). As data regarding particle size distribution and
emission factors from biomass burning in the Northern Territory of Australia are still
limited, and data for most states of Australia are unavailable, the impact assessments
of biomass burning in these regions remains highly speculative.
235
6.2. PRINCIPAL SIGNIFICANCE OF FINDINGS
The research reported in this thesis significantly advances an understanding of the
biomass burning processes in Australia, as well as on a global scale The study
applied a number of strategies for investigating characteristics of biomass burning
particles. Specifically, it involved designing a measurement system, optimizing the
measurement system, characterizing biomass burning particles, and taking an
inventory of emission factors in laboratory and field measurements. Figure 6.1
describes the research activities in this study. The first step of this study was
designing a burning system to investigate the effects of burning conditions associated
with controlled air supply on the characteristics of particles. Then the burning system
was investigated and optimized for best performance. The next step was
characterization of particles emitted by burning several species of trees common to
Queensland forests under controlled conditions using the system. The outcome of
this investigation was published and the technique with the greatest potential for
advanced applications in characterization of biomass burning particles in other states
was identified. Characterization of particles from burning of grass taken from
savannas in the Northern Territory of Australia in a laboratory was the next step in
obtaining valuable knowledge of biomass burning particles in the region. Finally, the
airborne measurement data of particle concentrations from biomass burning in the
Northern Territory of Australia taken during research campaigns involving the
International Laboratory for Air Quality and Health, Queensland University of
Technology (ILAQH, QUT), the Defence Science and Technology Organisation
(DSTO), and the Australian Commonwealth Scientific and Industrial Research
Organization (CSIRO), were analyzed and interpreted in order to have a
236
comprehensive understanding of characteristics of biomass burning particles in the
specific region and to get a better understanding of biomass burning in general.
The outcomes of this research are the characterisation of biomass burning
particles and emission factors from biomass burning in the Northern Territory of
Australia and Queensland. The inventory of the characteristics of biomass burning
particles contributes valuable knowledge to assessing the impacts in these particular
regions and provides a better understanding of biomass burning in general. These
inventories of emission factors are essential for modeling of source particle
production from biomass burning, which is needed for a particle dispersion model in
the atmosphere. The emission factors are very important in modeling human
exposures.
Laboratory study Field study
Airborne measurements in the Northern Territory of Australia during dry seasons
Particle characteristics and emission factors
Investigation of particle characteristics and emission factors from burning vegetations common to the Northern Territory of Australia and Queensland
Analysis and interpreting the airborne data
Designing a measurement system for biomass burning particles and optimizing the system
Impact assessments
Dispersion model, human exposure model
Figure 6.1. Diagram of research activities. The dashed rectangle shows activities which were not a part of this study, and the double dashed rectangle shows activities recommended for the future based on these results.
237
The principal findings and significance of this study are summarized as follows:
1. A new technique has been developed for characterizing biomass burning
particles under laboratory conditions closely simulating the environmental
conditions, which was an important step towards investigating biomass
burning processes in the laboratory. Investigation of the impacts of burning
rate associated with various wind speeds on the characteristics of biomass
burning particles in the field is not a trivial task and thus has never been
previously conducted by any researchers. The difficulty arising from the fact
that uncontrolled biomass burning may occur with a variety of burning rates.
In addition, the measurement system was designed taking into consideration
the facts that the concentrations of biomass burning particles are typically
very high, whilst the available instrumentation enables monitoring of a
limited range of particle concentrations, also biomass burning particles
contain chemical compounds that combine with water or other chemical
compounds to become larger particles. By taking into account these factors,
the measurement system consisted of three main parts: a modified stove used
to simulate burning conditions, by injecting air with the speed of 20 m/s (72
km/h) (a typical wind speed of bush fires in Australia) into the stove for “fast
burning”, and by keeping the stove unconnected to the blower during “slow
burning”, so as not to force the air supply through the ventilation system
(unforced flow rate of the incoming air was about 2 m/s); a diluter used to
dilute the smoke samples to a measurable concentration; and a particle
measurement set up. Optimisation of the system was undertaken in order to
obtain high performance. The performance of the measurement system was
238
determined by adjusting the flow rate of air in the dilution tunnel and the
temperature of the heated air in the Dekati diluter, and was found to be the
best when the air flow rate in the dilution tunnel and the injected air
temperature were 1 m/s and 200 oC, respectively. The stability of the systems
performance was demonstrated by the stability of the dilution ratio measured
during the experiments.
2. An important finding of this research is that burning conditions (such as fast
burning or slow burning) significantly influence the characteristics of particle
emissions. This research revealed a significant correlation between burning
condition and particle size distribution, particle number and mass emission
factor. It was also found that fast burning produces a large number of small
particles, while slow burning results in generating larger particles, though
fewer in number.
3. The research found that phase of burning affects the characteristics of
particles arising from biomass burning. For wood burning, more particles
with a larger diameter are emitted during the ignition and flaming phase,
while fewer and smaller particles are produced during the smoldering phase.
In the case of grass burning, the trend of the particle characteristic in every
phase was found to significantly depend on the species of grass.
4. Another finding of this research is that particles emitted during the burning of
samples from trees common to the Queensland forests, are found in a specific
count median diameter (CMD). Fast burning of the wood samples produced
particles with the CMD of 60 nm during the ignition phase and 30 nm for the
rest of the burning process. Slow burning of the wood samples releases large
239
particles with the CMD of 120 nm, 60 nm and 40 nm for the ignition, flaming
and smoldering phases, respectively. The CMD of particles emitted by the
burning of leaves and branches, were found to be 50 nm for the flaming and
30 nm for the smoldering, under fast burning conditions. Under slow burning
conditions, the CMD of particles was found to be between 100 to 200 nm for
the ignition and flaming phase, and 50 nm for the smoldering phase.
5. This research is mainly focused on quantifying the particle number emission
factor from biomass burning as they are recognized as important parameters
in assessing biomass burning impacts. Important findings were that the
particle number emission factor from burning of Queensland trees depends on
species of tree, part of tree burnt, and burning condition. Fast burning emits
particles with number emission factor ranging from 3.3 - 5.7 x 1015
particles/kg for woods, and 0.5 - 6.9 x 1015 particles/kg for leaves and
branches. Slow burning produces particle number emission factor in the range
of 2.8 - 44.8 x 1013 particles/kg for woods, and 0.5 - 9.3 x 1013 particles/kg
for leaves and branches.
6. The quantification of emission factor of PM2.5 from burning trees commonly
found growing in Queensland forests is another finding of this research. The
PM2.5 emission factor from wood burning is found in the range of 140 to 210
mg/kg and the PM2.5 emission factor from leaf and branch burning and grass
burning is 450 – 2000 mg/kg. The important finding in this research is that
the PM2.5 emission factor is not only influenced by the type of vegetation
burnt, but also affected by burning condition.
240
7. Further important findings are the characteristics of particle size distribution
from burning grasses predominantly growing in savannah regions of the
Northern Territory of Australia under different burning conditions. The
majority of particles were found with a CMD to be 50 nm for fast burning,
and 165 nm for slow burning. The particle emission factors and PM2.5
emission factors from the biomass burning are other important outputs of this
study. The number emission factor and PM2.5 emission factor are in the
ranges of 2.3 x 1014 - 7.5 x 1015 particles/kg and 450 - 4500 mg/kg
respectively. The burning condition, phase of burning, and species of grass
are found to be the factors significantly determining the characteristics of
particles.
8. The findings for the characteristics of particle size distribution, particle
number emission factor and PM2.5 emission factor are important when
assessing the characteristics of biomass burning in other states of Australia,
where the vegetation is similar to that growing in Queensland and the
Northern Territory of Australia. The eucalypt trees, commonly found growing
in Queensland forests, are similar to the trees found in other states of
Australia, which experience biomass burning, including the Northern
Territory of Australia, New South Wales and Victoria. The grasses
commonly found in the savannas in the Northern Territory of Australia are
also commonly found in other savanna regions in other states of Australia.
9. The reported data show that biomass burning in the Northern Territory of
Australia mostly occurs during the dry season. The characteristics of biomass
burning particles in the dry season are important issues in understanding their
241
impacts. This research revealed differences in particle characteristics during
early and late dry season due to different environmental conditions, including
moisture content of vegetation (less moisture content in the late dry season),
intensity of fires (higher intensity of fires in the late dry season), location of
fires related to the fight paths (the fires were located far from the flight paths
for the measurements conducted in the late dry season) and the size of burned
areas (larger area burned in the late dry season). Small particles in low
concentration were found in the early dry season, while large particles in high
concentration were found in the late dry season. The majority of particles
found under the boundary layer had an average diameter of 83 nm in the early
dry season, and particles with an average diameter of 127 nm were found in
the late dry season.
10. The research found similar characteristic particle size distributions from
biomass burnings measured both under controlled conditions in a laboratory
and in the field. The quantification of size distribution of particles emitted by
burning grasses from the Northern Territory of Australia under laboratory
controlled conditions with the burning rate varying from 2 m/s to 20 m/s
found that the majority of the particles were within a CMD in the range of 30
to 210 nm, depending on species of grass and burning phase. The field
measurements conducted with the wind speed of 5 m/s, recorded particles
under the boundary layer with CMD of 83 nm for the early dry season and of
127 nm for the late dry season.
11. Finding a vertical profile of biomass burning particles in the atmosphere over
the Northern Territory of Australia is a further important contribution to this
242
research. Investigation of the vertical profile of particles over a range of
height levels provided valuable insights on the characteristics of particles in
the atmosphere. These data are also of significance for modeling particle
dispersion and undertaking impact assessments. The present research found
that the characteristics of the size distribution of particles significantly
influence the vertical profile of particles. Characteristics of particles
remaining under the boundary layer are found to be dependent on biomass
burning particles released in the region at the time. Profiles of particles
penetrating the boundary layer and remaining above the boundary layer are
found to be dependant on the characteristics of particles coming from under
the boundary layer. Most particles remaining in the free troposphere are
existing particles and particles from other sources and depend on season.
12. Overall, the findings of this study significantly contribute towards advancing
a scientific understanding of biomass burning processes and provide valuable
tools for different aspects of impact assessment studies. The quantification of
emission factors in two states of Australia, Queensland and the Northern
Territory, that frequently experience biomass burning are primary findings of
this research. These data are very important for modeling of dispersion of
biomass burning particles from their sources to receptors and essential for
estimating the dose of biomass burning particles received by humans living in
the regions, which is of significance for the impact assessments. These
findings are particularly important for these two states (Queensland and the
Northern Territory), but also for Australia in general.
243
6.3. CONLUSIONS
Comprehensive studies, involving both laboratory investigations and field
measurements of biomass burning, were conducted in order to investigate the many
factors that influence biomass burning particles and their characteristics. A number
of results were found in this study, for the different measurement conditions, which
have helped build a better understanding of the characteristics of biomass burning
particles, both in general and for Australia in particular.
6.4. SCIENTIFIC RECOMMENDATIONS
Because of the complexity of factors and processes influencing the characteristics of
biomass burning particles that occur both during particle production and particle
distribution in the atmosphere, there are many challenges in developing a better
understanding of biomass burning. There are a number of recommendations for
future studies:
1. Development of method. The method for characterising biomass burning
particles in a laboratory needs to be further developed in future by taking into
account other factors in the field that affect the characteristics of particles.
2. Analysis of chemical compounds. Because biomass burning involves complex
physical and chemical processes in particle production, it is very important to
investigate the chemical compounds produced by the different burning
conditions carried out in this research. Chemical analysis of particles is very
important for identifying the major chemical compounds of biomass burning
particles and for advancing understanding of particle characteristics, as well as
244
the processes influencing particle characteristics. It is, therefore, recommended
that future research be focused on these aspects.
3. Flight path of airborne measurements. The airborne measurements were
carried out on a single, linear flight path. As a result, two dimensional vertical
profiles of particles were obtained in this research. It is therefore recommended
to conduct measurements over two intersecting flight paths, in order to obtain a
vertical profile of particles in three dimensions. This research would also be
useful for three dimensional dispersion model.
4. Future field study. It is recommended to conduct research aimed at estimating
the magnitude of burnt areas. This information is needed to calculate the total
amount of biomass burning in the area. The combined data of the total area of
biomass burnt and the emission factors are required to estimate the total
particles released into the atmosphere during the period of burning. The data
are also needed for modeling dispersion of biomass burning particles in the
atmosphere.
5. Modeling. The inventory of particle emission factors in this study is essential
data for modeling several processes:
a. Source strength model. A source strength model is related to the
emission production from a source. The model needs the data of emission
factors, type of emission, and rate of released heat.
b. Dispersion model for biomass burning particles in the atmosphere. A
dispersion model of particles from biomass burning includes a source
strength model and a meteorological model as the input.
245
c. Exposure of biomass burning particles at the receptors. Quantifying
the exposure of biomass burning particles to humans is necessary to
evaluate health impacts. Human exposure to biomass burning particles
involves quantifying the dose of particles received in a certain time. The
data of particle emission factors from a source is thus needed to estimate
the distribution of the concentration as a function of distance and time.
The particle exposure of biomass burning particles to humans living in a
certain distance from the source and at a certain time can be modeled.
6.4. REFERENCES:
Anderson, B. E., W. B. Grant, G. L. Gregory, E. V. Browell, J. E. Collins Jr, D. W.
Sachse, D. R. Bagwell, C. H. Hudgins, D. R. Blake and N. J. Blake,(1996),
Aerosols from biomass burning over the tropical South Atlantic region:
Distributions and impacts. Journal Geophysical Research, 101,24117-24137.
Andreae, M. O.,(1991), In Global Biomass Burning: Atmospheric Climatic and
Biospheric Implications; Levine, J.S.E., The MIT Press, Cambridge, MA.
Andreae, M. O., T. W. Andreae, H. Annegarn, J. Beer, H. Cachier, P. Le Canut, W.
Elbert, W. Maenhaut, I. Salma, F. G. Wienhold and T. Zenker,(1998),
Airborne studies of aerosol emissions from savanna fires in southern Africa:
246
2. Aerosol chemical composition. Journal Geophysical Research,
103,32119-32128.
Andreae, M. O., E. V. Browell, M. Garstang, G. L. Gregory, R. C. Harris, G. F. Hill,
D. J. Jacob, M. C. Pereira, G. W. Sachse, A. W. Setzer, P. L. Silvia Dias, R.
W. Talbot, A. L. Torres and S. C. Wolfsy,(1988), Biomass burning emissions
and associated haze laser over Amazonia. Journal of Geophysical Research,
93,1509-1527.
Areskoug, H., P. Camner, S. E. Dahlén, L. Låstbom, F. Nyberg, G. E. Pershagen and
A. Sydbom,(2000), Particles in ambient air - a health risk assessment.
Scandinavian Journal of Work, Environment and Health, 26,1-96.
Bodhaine, B. A.,(1983), Aerosol measurements at four background sites. Journal of
Geophysical Research, 88,10753 - 10768.
Cahoon, D. R., B. J. Stock, J. S. Levine, W. R. Cofer III and C. C. Chung,(1992),
Evaluation of a technique for satellite-derived estimation of biomass burning.
Journal of Geophysical Research, 97(D4),3805-3814.
Chalupa, D. C., P. E. Morrow, G. Oberdorster, M. J. Utell and M. W.
Frampton,(2004), Ultrafine particle deposition in subjects with asthma.
Environmental Health Perspective, 112,879-82.
Cofer III, W. R., E. L. Winstead, B. J. Stock, L. W. Overby, J. G. Goldammer, D. R.
Cahoon and J. S. Levine,(1996), Emissions from boreal forest fires: are the
atmospheric impacts underestimated ?, in: Global Biomass Burning and
Global Change, edited by Levine, J.S., MIT Press, Cambridge, Mass, pp 716-
7332
247
Delfino, R. J., C. Sioutas and S. Malik,(2005), Potential role of ultrfine particles in
associations between airborne particle mass and cardiovascular health.,
Environmental Health Perspective, 113(8),934-946.
Dennis, A., M. Fraser, S. Anderson and D. Allen,(2002), Air pollutant emissions
associated with forest, grassland, and agricultural burning in Texas.
Atmospheric Environment, 36(23),3779-3792.
Dhalla, N. S., R. M. Temsah and T. Netticadan,(2000), Role of oxidative stress in
cardiovascular health., J. Hypertens, 18,655-673.
Dubovik, O., B. N. Holben, T. F. Eck, A. Smirnov, Y. J. Kauffman, M. D. King, D.
Tanre and I. Slutsker,(2002), Variability of absorption and optical properties
of key aerosol types observed in worldwide locations. Journal Atmospheric
.Science, 59,590-608.
Eck, T. F., B. N. Holben, J. S. Reid, N. T. O'Neill, J. S. Schaller, O. Dubovik, A.
Smirnov and M. A. Yamasoe,(2003), High aerosol optical depth biomass
burning avents: a comparasion of optical properties for different source
regions. Geophysical Research Letters, 30(24),2293, doi:
10.1029/2003GL018697.
Elguren-Fernandez, A., A. H. Miguel, P. Jaques and C. Sioutas,(2003), Evaluation of
a denuder MOUDI-PUF sampling system to determine the size distribution of
semivolatile polycyclic aromatic hydrocarbons in the atmosphere. Aerosol
Science and Technology, 37,201-209.
Etzel, R.,(1999), A research highlights: Air pollution and bronchitis symptoms in
Southern California children with asthma. Environmental Health
Perspectives, 107(9)
248
Ferge, T., J. Maguhn, K. Hafner, F. Muhlberger, M. Davidovic, R. Warnecke and R.
Zimmermann,(2005), On-line analysis of gas phase composition in the
combustion chamber and particle characteristics during combustion of wood
and waste in a small batch reactor. Environmental Science and Technology,
39(6),1393-1402.
Fiebig, M., A. Stohl, M. Wendisch, S. Eckhardt and A. Petzold,(2003), Dependence
of solar radioactive forcing of forest fire aerosol on aging and state of
mixture. Atmospheric Chemistry and Physics, 3,881-891.
Fine, P. M., G. R. Cass and B. R. T. Simoneit,(2001), Chemical characterization of
fine particle emissions from fireplace combustion of woods grown in the
Notheastern United States. Environmental Science and Technology, 35,2665-
2675.
Fine, P. M., G. R. Cass and B. R. T. Simoneit,(2002), Chemical characterization of
fine particle emissions from the fireplace combustion of woods grown in the
Southern United States. Environmental Science and Technology, 36,1442-
1451.
Formenti, P., W. Elbert, W. Maenhaut, J. Haywood, S. Osborne and M. O.
Andreae,(2003), Inorganic and carbonaceous aerosols during the Southern
African Regional Science Initiative (SAFARI 2000) experiment: Chemical
characteristics, physical properties, and emission data for smoke from African
biomass burning. Journal of Geophysical Research, 103(D13),8488,
doi:10.1029/2002JD002408.
Formenti, P., T. Reiner, D. Sprung, M. O. Andreae, M. Wendisch, H. Wex, D.
Kindred, K. Dewey, J. Kent, M. Tzortziou, A. Vasaras and C. Zerefos,(2002),
249
STAAARTE-MED 1998 summer airborne measurements over the Aegean
Sea: 2. Aerosol scattering and absorption, and radiative calculations. Journal
of Geophysical Research, 107,doi:10.1029/2001JD001536.
Gao, S., D. A. Hegg, P. V. Hobbs, T. W. Kirchstetter, B. I. Magi and M.
Sadilek,(2003), Water soluble organic components in aerosols associated
with savanna fires in southern Africa: identification, evolution, and
distribution., Journal of Geophysical Research, 108(D13),8491, doi:
10.1029/2002JD002324.
Glasmann, I.,(1988), Soot formation in combustion processes. In Twenty-Second
Symposium (International) on combustion, The combustion Institute,
Pittsburgh, p.295.
Goldberg, M. S., J. C. I. Bailar, R. T. Burnett, J. R. Brook, R. Tamblyn, Y. Bonvalot,
P. Ernst, K. M. Flegel, R. K. Singh and M.-F. Valois,(2000), Identifying
subgoups of the general population that may be susceptible to short-term
increases in particulate air pollution: A time series study in Montreal,
Quebec. Health Effects Institute, Research Report Number 97
Gras, J. L.,(1991), Southern hemisphere tropospheric aerosol microphysics., J.
Geophys. Res ,, 96(D3),5345-5365.
Gras, J. L.,(1999), Some optical properties of smoke aerosol in Indonesia and
Tropical Australia. Geophysical Research Letter, 26(10),1393-1396.
Hays, M. D., C. D. Geron, K. J. Linna, N. D. Smith and J. J. Schauer,(2002),
Speciation of gas-phase and fine particle emissions from burning of foliar
fuels. Environmental Science and Technology, 36,2281-2294.
250
Haywood, J. M., S. R. Osborne, P. N. Francis, A. Keil, P. Formenti, M. O. Andreae
and K. P. H.,(2003), The mean physical and optical properties of regional
haze dominated by biomass burning aerosol measured from the C-130 aircraft
during SAFARI 2000. Journal Geophysical Research, 108,doi:
10.1029/2002JD002226.
Hedberg, E., A. Kristensson, M. Ohlsson, C. Johansson, P.-A. Johansson, E.
Swietlicki, V. Vesely, U. Wideqvist and R. Westerholm,(2002), Chemical
and physical characterization of emissions from birch wood combustion in a
wood stove. Atmospheric Environment, 36(30),4823-4837.
Hobbs, P. V., J. S. Reid, J. A. Herring, J. D. Nance, R. E. Weiss, J. L. Ross, D. A.
Hegg, R. D. Ottmar and C. Liousse,(1996), Particle and gas measurements in
smoke from prescribed burns of forest products in the Pacific Northwest. in:
Biomass Burning and Global Change, Vol. 1. edited by Levine, J.S, pp. 697-
715, MIT Press, New York
Hueglin, C. H., C. H. Gaegauf, S. Kunzel and H. Burtscher,(1997), Characterization
of wood combustion particles: morphology, mobility, and photoelectric
activity. Environmental Science and Technology, 31,3439-3447.
Kaufman, Y. J., P. V. Hobbs, V. Kirchhoff, P. Artaxo, L. Remer, B. N. Holben, M.
D. King, D. E. Ward, E. M. Prins, K. M. Longo, L. F. Mattos, C. A. Nobre, J.
D. Spinhirne, J. Q. Thompson, A. M. Gleason, S. A. Christopher and S. C.
Tsay,(1998), Smoke, Clouds, and Radiation-Brazil (SCAR-B) experiment. J.
Geophys. Res-A,, 103,31783-31808.
Koppmann, R., K. V. Craplewski and J. S. Reid,(2005), A review of biomass burning
emissions, part I: gasesous emissions of carbon monoxide, methane, volatile
251
organic compounds, and nitrogen containing compounds. Atmospheric
Chemistry and Physics Discussion, 5,10455-10516.
Le Canut, P., M. O. Andreae, G. M. Harris, F. G. Weinhold and T. Zenker,(1996),
Airborne studies of emissions from savannas fire in southern Africa, 1,
Aerosol emissions measured with a laser optical particle counter. Journal of
Geophysical Research, 101,23615-23630.
Levine, J. S.,(1991), Global Biomass Burning: Atmospheric, Climatic, and
Biospheric Implication. MIT Press, Cambridge, Mass.
Li, N., C. Sioutas, J. R. Froines, A. Cho, C. Misra and A. Nel,(2003), Ultrafine
particle pollutants induce oxidative stress and mitochondrial damage.
Environmental Health Perspective, 111,455-460.
Martins, J. V., P. Artaxo, C. Liousse, J. S. Reid, P. V. Hobbs and Y. J.
Kaufman,(1998), Effects of black carbon content, particle size, and mixing on
light absorption by aerosols from biomass burning in Brazil., J. Geophys.
Res-A,, 103,32041-32050.
McConnel, R., K. Berhane, F. Gilliland, S. J. London, H. Vora, E. Avol, W. J.
Gauderman, H. G. Margolis, F. Lurmann, D. C. Thomas and J. M.
Peters,(1999), air pollution and bronchitis symptoms in Southern California
Children with asthma. Environmental Health Perspectives, 107,757-760.
McDonald, J. D., B. Zielinska, E. M. Fujita, J. C. Sagebiel, J. C. Chow and J. G.
Watson,(2000), Fine particle and gaseous emission rates from residential
wood combustion. Environmental Science and Technology, 34,2080-2091.
252
Nel, A., D. Diaz-Sanchez, D. Ng, T. Hiura and A. Saxon,(1998), Enhancement of
allergic inflammation by the interaction between diesel exhaust particles and
the immune system. Journal of Allergy Clinical Immunology, 102,539-554.
Nel, A. E., D. Diaz-Sanchez and N. Li,(2001), The role of particulate pollutants in
pulmonary inflammation and asthma: evidence for the involvement of
organic chemicals and oxidative stress., Curr Opin Pulm Med, 7,20-26.
Nemmar, A., P. H. M. Hoet, M. Thomeer, B. Nemery, B. Vanquickenborne and H.
Vanbilloen,(2002), Passage of inhaled particles into the blood circulation in
humans. Circulation, 105,411-414.
Nemmar, A., M. F. Hoylaetr, P. H. M. Hoet and B. Nemery,(2004), Possible
mechanisms of cardiovascular effects oh inhaled particles: systemic
translocation and prothrombotic effects. Toxicology Letter, 149,243-253.
Nichol, J.,(1997), Bioclimatic impacts of the 1994 smoke haze event in southeast
Asia. Atmospheric Environment, 31(8),1209-1219.
Norris, G., S. N. YoungPong, T. V. Larson and J. W. Stout,(1999), An association
between fine particles and asthma emergency department visits for children
in Seatle. Environmental Health Perspectives, 107,489-493.
Oberdorster, G.,(2001), Pulmonary effects on inhaled ultrafine particles. Int Arch
Occup Environ Health, 74,1-8.
Oberdorster, G., Z. Sharp, V. Atudorei, A. Elder, R. M. Gelein and A. Lunts,(2002),
Extrapulmonary translocation of ultrafine carbon particles following whole
body inhilation exposure of rats. J. Toxicology Environmental Health A,
65,1531-1543.
253
Oberdorster, O. G., R. M. Gelein, J. Ferin and B. Weiss,(1995), Association of
particulate air pollution and acute mortality: involvement of ultrafine
particles?, Inhalation Toxicology, 7,111-124.
Ortiz de Zarate, I., A. Ezcurra, J. P. Lacaux and P. Van Dinh,(2000), Emission factor
estimates of cereal waste burning in Spain. Atmospheric Environment,
34(19),3183-3193.
Pekkanen, J., K. L. Timonen, J. Ruuskanen, A. Responen and A. Mirme,(1997),
Effects of ultrafine and fine particles in urban air on peak flow espiratory
flow among children with asthmatic symptoms. Environmental Research,
74,24-33.
Peters, A., E. Liu, R. L. Verrier, J. Schwartz, D. R. Gold, M. Mittleman, J. Baliff, J.
A. Oh, G. Allen, K. Monahan and D. W. Dockery,(2000), Air pollution and
incidence of cardiac arrhythmia. Epidemiology, 11(1),11-17.
Peters, A., S. Perz, A. Doring, J. Stieber, W. Koenig and H. E. Wichmann,(1999),
Increase in heart rate during an air pollution episode. American Journal of
Epidemiology, 150,1094-1098.
Peters, A., H. E. Wichmann, T. Tuch, J. Heinrich and J. Heyder,(1997), Respiratory
effects are associated with the number of ultra fine particles., American
Journal Respiratory Critical Care Medicine, 155,1376-1383.
Peters, J. M., E. Evol, W. Navidi, S. J. London, W. J. Gauderman, F. Lurmann, W. S.
Linn, H. Margolis, E. Rappaport, J. Hong, Jr and D. C. Thomas,(1999), a
study of twelve southern California communitis with differing levels and
types of Air pollution: I. Prevalence of respiratory morbidity. American
Journal Respiratory Critical Care Medicine, 159,760-767.
254
Pope, C. A., M. J. Thun, M. M. Namboodri, D. W. Dockery, J. S. Evans, F. E.
Speizer and C. W. Heath,(1995), Particulate air pollution as a predictor of
mortality in a prospective study of U.S adults. American Journal of
Respiratory Critical Care Medicine, 95,669-674.
Pritchard, R. J., A. J. Ghio, J. R. Lehmann, D. W. Winsett, J. S. Tepper and P.
Park,(1996), Oxidant generation and lung injury after particulate air pollutant
exposure increase with the concentrations associated metals., Inhalation
Toxicology, 8,457-477.
Reid, J. S. and P. V. Hobbs,(1998), Physical and optical properties of smoke from
individual biomass fires in Brazil. Journal of Geophysical Research,
103,32013-32031.
Reid, J. S. and P. V. Hobbs,(1998), Physical and optical properties of smoke from
individual biomass fires in Brazil. Journal Geophysical Research,
103,32013-32031.
Reid, J. S., P. V. Hobbs, R. J. Ferek, D. Blake, J. V. Martins, M. R. Dunlap and C.
Liousse,(1998), Physical, chemical, and optical properties of regional hazes
dominated by smoke in Brazil. Journal of Geophysical Research, 103,32059-
32080.
Reid, J. S., R. Koppmann, T. F. Eck and D. P. Eleuterio,(2005), A review of biomass
burning emissions part II: intensive physical properties of biomass burning
particles. Atmospheric Chemistry and Physics, 5,799-825.
Seaton, A., A. Soutar, V. Crawford, R. Elton, S. McNerian and J. Cherrie,(1999),
Particulate air pollution and the blood. Thorax, 54,1027:1032.
Shaw, R. W.,(1987), Air pollution by particles. Science Environment, 255,96 - 103.
255
Tolbert, P. E., J. A. Mulholland, D. D. MacIntosh, F. Xu, D. Danniel, O. J. Devine,
B. P. Carlin, M. Klein, J. Dorley, A. J. Butler, D. F. Nordenberg, H. Frunkin,
P. B. Ryan and M. C. White,(2000), Air quality and pediatric emergency
room visit for asthma in Atlanta, Georgia. American Journal of
Epidemiology, 151,798-810.
Tsutsumi, Y.,(1999), Aircraft measurement of ozone, NOx, CO, and aerosol
concentrations in biomass burning smoke over Indonesia and Australia in
October 1997: Depleted ozone layer at low altitude over Indonesia.,
Geophysical Research Letters, 26(5),595-598.
Uherek, E.,(2004). Vegetation fire, Max Planck Institute for Chemistry, Mainz,
Accessed Mei 2004, http://www.atmosphere.mpg.de/enid/238.html.
Vedal, S., J. Petkau, R. White and J. Blair,(1998), Acute affects on ambient inhalable
particles in asthmatic and nonasthmatic children. American Journal
Respiratory Critical Care Medicine, 157(4),1034-1043.
Von Klot, S., G. Wolke, T. Tuch, J. Heinrich, J. Heyder, H. E. Wichmann and A.
Peters,(2000), Short term effects of ultrafine and fine particles on medication
use in asthmatic adults. Proc. Conf. American Thoracic Soc.,2000. Toronto
(abstract).
Wichmann, H. E. and A. Peters,(2000), Epidemiological evidence of the effects of
ultrafine particle exposure. Phil. Trans. R. Soc. Lond. A, 358,2751-2769.
Wichmann, H. E., C. Spix, T. Tuch, G. Wolke, A. Peters and J. Heinrich,(2000),
Dailiy mortality and fine and ultrafine particles in Erfurt, Germany. Part i:
Role of particle number and particle mass. Res Rep Health Eff Inst, 98,5-86.
256
Wieser, U. and C. k. Gaegauf,(2005). Nanoparticle emissions of wood combustion
processes, Laboratories for Sustainable Energy System, Accessed March
2005,
http://www.oekozentrum.ch/downloads/publikationen/nanoparticles.pdf.
Wurzler, S. and M. Simmel,(2005). Impact of vegetation fires on composition and
circulation of the atmosphere, Accessed March 2005,
http://projects.tropos.de:8088/afo200g3/.
Yu, O., L. Sheppard, T. Lumley, J. Q. Koenig and G. G. Shapiro,(2000), Effects of
ambient air pollution on symptoms of asthma in Seattle- area children in the
CAMP study. Environmental Health Perspectives, 108(12),1209-1214.
257