Emissions Estimation

Technique Manual
for

Aggregated Emissions from

Domestic Gaseous Fuel Burning

September 1999

Version 1.0 - First Published 1 September 1999

 EMISSIONS ESTIMATION TECHNIQUE MANUAL:
AGGREGATED EMISSIONS FROM DOMESTIC GASEOUS FUEL
BURNING

TABLE OF CONTENTS

1.0 Introduction ....................................................................................................1

1.1 The NPI ................................................................................................1
1.2 Purpose and Scope of the Manual ...............................................1
1.3 Application of the Manual...............................................................2

2.0 Emissions Covered by the Manual...........................................................4

2.1 NPI Substances .................................................................................4
2.2 Emission Sources and Related Processes ................................4
2.3 Emission Controls ............................................................................4

3.0 Emissions Estimation Techniques...........................................................5

3.1 Approaches Employed.....................................................................5
3.1.1 Determination of Gaseous Fuel Consumed....................................5
3.1.2 Determination of Emissions of NPI Substances.............................7
3.2 Spatial Surrogates and Spatial Allocation .................................7
3.3 Emission Factors ..............................................................................9
3.4 Sample Calculations...................................................................... 10

4.0 Uncertainty Analysis ................................................................................. 12

4.1 Data Reliability ................................................................................ 12
4.2 Reliability of Emission Factors................................................... 12
4.3 Problems and Issues Encountered ........................................... 12
4.4 Recommendations for Further Work ........................................ 13

5.0 Glossary of Terms and Abbreviations.................................................. 14

6.0 References................................................................................................... 15

i
 EMISSIONS ESTIMATION TECHNIQUE MANUAL:
AGGREGATED EMISSIONS FROM DOMESTIC GASEOUS FUEL
BURNING

LIST OF TABLES

Table 1: NPI Substances Emitted by Domestic Gaseous Fuel Burning.....4

Table 2: Typical Energy Contents of Gaseous Fuels around Australia .....6
Table 3: Emission Factors for Domestic Gaseous Fuel Burning.................9
Table 4: Sample Energy Consumption Data................................................... 10
Table 5: Sample Data on Town Gas Usage..................................................... 10
Table 6: Sample Data for Notional Postcode Regions................................. 10

ii
1.0 Introduction

1.1 The NPI

The National Pollutant Inventory (NPI) was established under a National

Environment Protection Measure (NEPM) made by the National Environment
Protection Council (NEPC) under Commonwealth, State and Territory
legislation on 27 February 1998. This Measure is to be implemented
progressively through the laws and administrative arrangements of each of
these participating jurisdictions (i.e. State and Territory Governments).

The NEPM and an associated Memorandum of Understanding for the NPI,

which have been published as a single document by the NEPC, provide more
details on the purpose and structure of the NPI, and the arrangements for
implementation of the NEPM that have been agreed by the jurisdictions.
Users of this Manual should read this publication if they are unfamiliar with the
NEPM or the NPI.

1.2 Purpose and Scope of the Manual

The NPI will be developed as an internet database designed to provide

information on the types and amounts of certain chemical substances being
emitted to the air, land and water environments. If the NPI is to achieve its
aim of communicating useful and reliable information to the community,
industry and governments on pollutants present in our environment, the
emissions estimation techniques (EETs) used to generate inputs to the NPI
need to be consistent, and the process for developing these techniques needs
to be transparent. This Manual has been developed, reviewed and finalised
in this context.

The NEPM contains a list of substances for which emissions will be reported
on an annual basis to the Commonwealth Government, which will then
compile and publish the NPI. The aggregated emissions manuals, of which
this is one, have been prepared to assist State and Territory Governments in
preparing these submissions, and to facilitate consistent reporting between
these jurisdictions.

State and Territory Governments will also be compiling and submitting

emissions data based on annual inputs from reporting facilities. These
facilities are primarily industrial enterprises which use (or handle, manufacture
or process) more than specified amounts of certain polluting substances, burn
more than specified amounts of fuel, or consume more than certain amounts
of energy. These amounts or “thresholds” (which are clearly defined in the
NEPM) govern whether an industrial facility is required to report and what
substances it is required to report on, and industry handbooks are being
developed to help industries to prepare the information for these reports.

The aggregated emissions manuals complement these handbooks, and are

intended to enable Governments to estimate emissions from non-industrial
activities (e.g. transportation, domestic and commercial activities) and

1
emissions from industry which are not reported because the relevant
thresholds are not exceeded.

Annual submissions are also to be prepared and submitted in conformance

with the NPI Data Model and Data Transfer Protocol. For emissions to the air
environment, this Protocol only requires jurisdictions to submit data on
emissions into the particular airsheds that are listed in the Protocol, and not to
the rest of each jurisdictional area. For example, in Victoria, emissions data
are only required for the Port Phillip and Latrobe Valley Regions. In addition,
emissions data are required to be submitted on a gridded basis, with each
jurisdiction determining a grid domain and grid cell size appropriate to its
needs and responsibilities (e.g. for air quality modelling purposes).

Therefore, in addition to recommending and providing details and examples of

appropriate emissions estimation techniques (EETs) for the relevant NPI
substances, this Manual provides guidance on the spatial allocation of
emissions and the use of area-based surrogates for accurately distributing the
activities or sources in question.

1.3 Application of the Manual

Each of the aggregated emissions manuals provides details of:

• the NPI substances that are expected to be emitted from the relevant
aggregated source type;
• the origins or sources of the emissions, and the processes that may
generate them;
• the impacts of any control equipment or procedures on those emissions;
• the broad approaches that may be employed in the estimation and spatial
allocation of emissions;
• details of emission factors to be used in the estimation of emissions; and
• a series of illustrative sample calculations for each estimation technique.

Each of the manuals also contains a section on “Uncertainty Analysis”, which

provides information and guidance to users on the reliability of the various
estimation techniques, problems and issues associated with their
development and application, and recommendations for their improvement. In
preparing the aggregated emissions manuals it has been recognised that
some jurisdictions already undertake detailed emissions inventories on a
regular basis, based on relatively sophisticated methodologies. For these
jurisdictions the manuals offer techniques which represent commonly
available best practice for emissions estimation in Australia (i.e. techniques of
high quality which can be employed by larger or more experienced
jurisdictions with an acceptable expenditure of time and effort). The most
recent developments in inventory methodology in Australia and overseas
have been considered in selecting and documenting these techniques.

Where a more simplified methodology for emissions estimation of acceptable

quality is available, it is recommended in the manual for the use of those
jurisdictions which may, for the time being at least, lack the data, resources or

2
expertise to employ a more sophisticated approach, or not see the need for
highly reliable estimates in that particular part of the inventory.

3
2.0 Emissions Covered by the Manual

This manual provides guidance on the estimation of aggregated emissions

from the domestic burning of gaseous fuels, including natural gas, town gas
and liquefied petroleum gas (LPG). As these fuels are widely used
throughout the country in domestic applications, their consumption can make
a significant contribution to regional emissions.

2.1 NPI Substances

Table 1 lists those substances in Table 2 to Annex A of the NEPM which are
emitted by domestic gaseous fuel burning.

Table 1: NPI Substances Emitted by Domestic Gaseous Fuel Burning.

Arsenic and compounds Manganese and compounds
Benzene Mercury and compounds
Beryllium and compounds Nickel and compounds
Cadmium and compounds Oxides of nitrogen
Chromium (VI) compounds Particulate matter ≤10µm (PM10)
Carbon monoxide Polycyclic aromatic hydrocarbons
Cobalt and compounds Selenium and compounds
Copper and compounds Sulphur dioxide
Cyclohexane Toluene
Formaldehyde Total volatile organic compounds
n-Hexane Zinc and compounds
Lead and compounds

2.2 Emission Sources and Related Processes

Domestic gaseous fuel burning is undertaken for cooking, space heating and
hot water heating. Natural gas and town gas supplies are limited by the
extent of the reticulation system whilst LPG for domestic applications
(primarily propane) is supplied in the form of large, portable gas bottles.

Town gas is usually supplied by a reticulated gas system that is not

connected to a larger regional or statewide natural gas supply. Town gas
systems often serve only a small area around a population centre and may
utilise a number of different types of gases.

Emissions are dependent on the amount and type of the gaseous fuel burnt,
and also on the temperature and efficiency of combustion, but these latter
influencing factors are not directly considered in the EETs.

2.3 Emission Controls

There are no emission controls to factor into the EETs.

Industrial gaseous fuel burning practices often incorporate emission controls

such as flue gas recirculation and special burners to control the emission of
oxides of nitrogen. However, this technology is not used in domestic
applications and is not considered in the EETs.

4
3.0 Emissions Estimation Techniques

The information required to estimate emissions from domestic gaseous fuel

burning is as follows:
• the amount of each fuel type burnt in domestic applications in the airshed
or jurisdiction; and
• the density or energy content of each fuel type.

3.1 Approaches Employed

3.1.1 Determination of Gaseous Fuel Consumed

Fuel Consumption

Information on domestic fuel consumption may be available from local natural

gas, town gas and LPG distributors in an airshed or jurisdiction. It is desirable
to contact all gas distribution companies that operate within the airshed of
interest to ensure that all gas consumption is accounted for in estimating
emissions.

If domestic gas usage data are not available from local gas distributors,
information on the amounts of natural gas, town gas and LPG consumed in
domestic appliances across individual jurisdictions is available from the
Australian Bureau of Agriculture and Resource Economics (ABARE).
Alternatively, data for domestic gas use may be sought from the Australian
Gas Association (AGA).

Data from both of these alternative sources must be scaled down from
jurisdiction to airshed level, based on the population fraction and taking into
account the availability of each fuel type in the airshed. If information on the
availability of fuel types is not obtainable, simple scaling according to
population should be undertaken.

Energy Contents

Gas consumption data should be requested in volume units for the estimation
of emissions. If usage data are provided in tonnes, the fuel density should be
requested to converted it to a volume basis. If information on gas use is
reported as petajoules (PJ), it must also be converted to volume.

If data on local energy contents cannot be obtained, Table 2 lists typical

energy contents of fuels used in Australia.

The types and amounts of town gas used in an airshed should be available
from local gas distributors. Alternatively, a breakdown of the proportion of
each type of town gas used in various jurisdictions is available from ABARE.

Once the total annual domestic energy consumption and energy content of
each type of fuel used within an airshed are known, the total volume of each
fuel consumed for domestic purposes can be calculated from Equation 1.

5
Table 2: Typical Energy Content of Gaseous Fuels around Australia
Type of Fuel Energy Contenta

Natural Gas
Victoria 38.8 MJ m-3
Queensland 38.5 MJ m-3
South Australia and New South Wales 38.9 MJ m-3
Western Australia 39.6 MJ m-3
Northern Territory 40.7 MJ m-3

Town gas
Synthetic natural gas 39.0 MJ m-3
Reformed gas 20.0 MJ m-3
Tempered LPGb 25.0 MJ m-3
Tempered natural gas 25.0 MJ m-3

LPG 25.3 MJ L -1
a
ABARE (1999).
b
Tempered LPG is supplied via a reticulation system in the form of a gas, and
should not be confused with LPG which is supplied as a bottled liquid.

Equation 1: Calculation of annual volumes of gaseous fuels burnt in

domestic applications in an airshed (based on energy consumption
data)

Vi = E i / (Q i * F i)

where

Vi = Volume of fuel type i consumed domestically in an

airshed, 106 m3 yr-1 (or 103 L for LPG)
Ei = Domestic energy consumption for fuel type i in an
airshed, PJ yr-1
Qi = Energy content of fuel type i, MJ m-3 (or MJ L -1 for LPG)
Fi = Conversion factor: 10-3 (for natural gas and town gas) or
10-6 (for LPG)

If specific information on the use of each form of town gas within an airshed is
not known, then ABARE information on the typical proportions of usage for
each town gas type in a jurisdiction should be used. From these typical
proportions, an estimate of the total volume of town gas used can be
calculated using Equation 2.

6
Equation 2: Calculation of annual volume of town gas burnt in domestic
applications in an airshed (where specific usage data are not available)

VTG = E TG * 10 3 * ∑ (PTGi Q TGi )

i

where

VTG = Total annual volume of town gas used in an airshed,

106 m3 yr-1
ETG = Total domestic energy consumption for town gas in an
airshed, PJ yr-1
QTGi = Energy content of town gas type i, MJ m-3
PTGi = Proportion of town gas usage across the jurisdiction for
town gas type i

3.1.2 Determination of Emissions of NPI Substances

Once the volume of each fuel burnt in domestic applications in an airshed has
been estimated, the annual emissions of each NPI substance for that fuel type
can be calculated using Equation 3.

Emission factors for town gas are not available and are assumed to be the
same as for natural gas.

Equation 3: Calculation of annual airshed emissions of an NPI

substance from combustion of particular fuel type

Eij = EFij * V i

where

Eij = Total annual emissions of substance j in the airshed

from fuel type i, kg yr -1
EFij = Emission factor for substance j from fuel type i,
kg 10-6 m-3 (or kg 10-3 L-1 for LPG)
Vi = Volume of fuel type i consumed annually in airshed,
106 m3 yr-1 (or 103 L yr-1 for LPG)

3.2 Spatial Surrogates and Spatial Allocation

Supply of natural gas and town gas is reliant on reticulation systems which
have definite boundaries. Annual emissions from use of these fuels should
therefore be spatially allocated according to the population distribution within
these reticulation areas.

7
If gas usage data can be acquired on the basis of postcode or some other
area basis, it is a relatively easy task to allocate emissions to their correct grid
cells (see Equation 4 below). If ABARE data are used, it is necessary to
obtain information on the boundaries of the reticulation areas from local gas
suppliers or the AGA, and to determine the postcodes (or other appropriate
area units) which receive gas supplies. If no spatial data are available, or if it
is known that the reticulation system covers the whole airshed, emissions
from natural gas and town gas can be allocated according to population alone
(see Equation 5).

LPG sales data are generally available from distribution companies by

postcode, which can be used for spatial allocation of emissions. If this
information is not available, LPG emissions can also be spatially allocated
according to the population within the airshed.

The Australian Bureau of Statistics collects population data by Collection

District (CD). This data can be converted to population by grid cells using a
program or Geographic Information System (GIS). If spatial distribution
information is provided by gas suppliers on a postcode basis, emissions in a
grid cell can be calculated using Equation 4. Use of a GIS is recommended in
applying Equation 4, as it can accurately calculate the proportion of a
postcode region’s population that is present in a grid cell.

Equation 4: Estimating emissions of an NPI substance in a grid cell for a

fuel type (based on gas usage data provided on a postcode basis)

E ijk = (Eij Vi ) * ∑ (Pkp Pp ) * Vip

p

where

Eijk = Annual emissions of substance j from fuel type i in grid

cell k, kg yr-1
Eij = Total annual airshed emissions of substance j from fuel
type i, kg yr-1
Pkp = Population of postcode region p in grid cell k
Pp = Total population of postcode region p
Vip = Volume of fuel type i used in postcode region p,
106 m3 yr-1 (or 103 L yr-1 for LPG)
Vi = Total volume of fuel type i use in the airshed, 106 m3 yr-1
(or 103 L yr-1 for LPG)

Energy consumption data for postcode regions and the airshed may be used
instead of data on fuel usage.

8
Equation 5: Estimating emissions of an NPI substance in a grid cell for a
fuel type (based on population alone)

Eijk = E ij * (P k / P)

where

Pk = Population in grid cell k

P = Total population of the airshed

3.3 Emission Factors

Emission factors for use in determining aggregate emissions from domestic
gaseous fuel burning are provided in Table 3.

Table 3: Emission Factors for Domestic Gaseous Fuel Burning

NPI Substance Natural Gasa LPGb
(kg 10-6 m-3) (kg 10-3 L-1)
-3
Arsenic and compounds 3.20 x 10
Benzene 3.36 x 10-2 2.66 x 10 –3 c
Beryllium and compounds 9.6 x 10-5 d
Cadmium and compounds 1.76 x 10-2
Chromium (VI) compounds 2.24 x 10-2 2.64 x 10-4 e
Carbon monoxide 640 0.228
Cobalt and compounds 1.34 x 10-3 9.60 x 10-4 e
Copper and compounds 1.36 x 10-2 2.40 x 10-5 e
Cyclohexane 1.74c 7.13 x 10-4 c
Formaldehyde 1.20 6.28 x 10-3 c
n-Hexane 28.8 7.38 x 10-4 c
Lead and compounds 8.0 x 10-3 2.40 x 10-5 e
-3
Manganese and compounds 6.08 x 10 2.40 x 10-5 e
Mercury and compounds 4.16 x 10-3
Nickel and compounds 3.36 x 10-2 2.64 x 10-4 e
Oxides of nitrogen 1500 1.68
Particulate matter ≤10µm 122 4.80 x 10-2
-2
Polycyclic aromatic hydrocarbons 1.1 x 10
Selenium and compounds 1.92 x 10-4 d 2.64 x 10-4 e
Sulphur dioxide 8.36f 9.45 x 10-9 g
-2
Toluene 5.4 x 10 1.34 x 10-3 c
Total volatile organic compounds 88.0 3.60 x 10-2
-1
Zinc and compounds 4.64 x 10 2.64 x 10-4 e
a
USEPA (1998) unless otherwise specified.
b
USEPA (1996) unless otherwise specified.
c
CARB (1991).
d
Emission factors based on 50% of USEPA (1998) method detection limit.
e
USEPA (1992).
f
Based on 2.09 times the typical Victorian sulfur content of 4 mg m -3 (EA 1999).
Indicative sulphur content data for other jurisdictions may also be used.

9
g
Based on 0.00021 times the Australian average sulfur content for LPG of 0.045 g
m -3 (EA 1999).

3.4 Sample Calculations

The data in Tables 4, 5 and 6 will be used in the sample calculations.

Table 4: Sample Energy Consumption Data

Energy Consumption in an Airshed by Fuel Type
Fuel Type Amount (PJ)
LPG 2
Natural Gas 25
Town Gas 3

Table 5: Sample Data on Town Gas Usage

Proportion of Different Types of Town Gas Used in a Jurisdiction
Town Gas Type Proportion (%)
Synthetic Natural Gas 90
Reformed Gas 10
Tempered LPG 0
Tempered Natural Gas 0

Table 6: Sample Data for Notional Postcode Regions

Postcode Population in Portion of Postcode Annual LPG Usage
Region Postcode Region Population in Postcode Region
Region in Grid Cell k (103 L yr-1)
P1 45 000 20 000 580
P2 38 000 8 000 220

Example 1: Calculation of annual volumes of gaseous fuels burnt in an

airshed (based on energy consumption data)

Using Equation 1 for natural gas and LPG, the sample data from Table 4, the
energy content from Table 2 for LPG, and an assumed energy content of
natural gas in the jurisdiction of 40.0 MJ m-3

Vi = E i / (Q i * F i)

Vnaturalgas = 25 / (40.0 *10-3)

= (6.25 * 102) 106 m3 yr-1

VLPG = 2 / (25.3 *10-6)

= (7.91 * 104) 103 L yr-1

For the following example involving town gas, Equation 2 is used because
usage data are only provided as an overall figure (not for each type of town
gas), but estimates of typical proportions of usage of each town gas type in
the jurisdiction are available.

10
Example 2: Calculation of annual volume of town gas burnt in an airshed
(where specific usage data are not available)

Using Equation 2, the data from Table 5, and the energy content data from
Table 2 for the two town gas types

VTG = E TG * 10 3 * ∑ (PTGi Q TGi )

i

VTG = 3 * 103 [0.90 / (39.0 * 10-3) + 0.10 / (20.0 * 10-3)]

= 84.2 106 m3 yr-1

Example 3: Calculation of annual airshed emissions of PM10 from

combustion of gaseous fuels

Using Equation 3, the emission factors from Table 3 (assuming the same EF
for town gas as natural gas), and the results of Examples 1 and 2

Eij = EFij * V i

Enaturalgas,PM10 = 122 * 6.25 * 102

= 7.63 * 104 kg yr -1

Etowngas,PM10 = 122 * 84.2

= 1.03 * 104 kg yr -1

ELPG,PM10 = 4.80 * 10-2 * 7.91 * 104

= 3.79 * 103 kg yr -1

Example 4: Estimating PM10 emissions in a grid cell from LPG usage

(based on usage data provided on a postcode basis)

Using Equation 4, the results for LPG from Examples 1 and 3, and the sample
data from Table 6

E ijk = (Eij Vi ) * ∑ (Pkp Pp ) * Vip

p

ELPG,PM10,cell = (E LPG,PM10 / V LPG) * [(P P1/cell / P P1) * V LPG,P1 +

(P P2/cell / P P2) * V LPG,P2 ]
= 3.79 * 103 / (7.91 * 104) * [(20000 / 45000) * 580 +
(8000 / 38000) * 220]
= 14.6 kg yr -1

11
4.0 Uncertainty Analysis

In the following analysis reliability is classified into 3 levels of confidence: high

(uncertainty of 20% or less), medium (uncertainty of between 20% and 80%)
and low (uncertainty of greater than 80%).

4.1 Data Reliability

Data on domestic gas usage from local gas distributors should be of high
reliability. ABARE and AGA data are considered to be of medium reliability as
they are based on gas consumption data provided by gas suppliers and LPG
distributors at a jurisdiction (rather than airshed) level.

4.2 Reliability of Emission Factors

Natural Gas Emission Factors

Most of the emission factors for natural gas combustion are based on AP-42
data (USEPA 1998) for uncontrolled residential gas burning. The factors for
CO, NOx , SO2 , benzene and formaldehyde are considered to be of high
reliability.

The remaining factors are considered to have medium reliability, with the
exception of factors for compounds of arsenic, beryllium, selenium, zinc and
polycyclic aromatic hydrocarbons which are considered to have low reliability.
The factor for cyclohexane is sourced from CARB and is also considered to
be of relatively low reliability.

Town Gas Emission Factors

There are no emission factors available for town gas, and the factors for
natural gas are recommended for use in this EET manual. As these emission
factors are not specific to each (or, indeed, any) form of town gas, the
reliability for all factors is considered to be lower than those for natural gas.
Thus, the factors for CO, NOx , SO2, benzene and formaldehyde are
considered to be of medium reliability, while the reliability of the remaining
factors is considered to be low.

LPG Emission Factors

The LPG emission factors for CO, NOx , SO2, VOC and PM10 are based on
the emission factors for uncontrolled commercial boilers in AP-42 (USEPA
1996). These factors are considered to be of medium reliability. The
remaining factors are based on speciation calculations by USEPA, and are
considered to have low reliability.

4.3 Problems and Issues Encountered

Data availability is the primary problem that limits the accuracy of the results
obtainable with these EETs. Most of the emission factors are based on
USEPA data and may not be very relevant to appliances used in Australia.

12
4.4 Recommendations for Further Work

Specific emissions tests on Australian appliances using Australian gaseous

fuels would allow more accurate emission factors to be developed.

Some manufacturers are developing the use of emission control technology

such as low NOx burners for application in domestic appliances. While this
technology is not common at this stage in Australia, its development should
be monitored as a significant change in technology will require a review of
emission factors.

13
5.0 Glossary of Terms and Abbreviations

ABARE Australian Bureau of Agricultural and Resource Economics

AGA Australian Gas Association
CARB California Air Resources Board
CO Carbon monoxide
EA Environment Australia
EET Emissions estimation technique
EPAV Environment Protection Authority of Victoria
GIS Geographic information system
LPG Liquefied petroleum gas
NEPC National Environment Protection Council
NEPM National Environment Protection Measure
NOx Oxides of nitrogen
NPI National Pollutant Inventory
PM10 Particulate matter less than or equal to 10 µm
PJ Petajoules
SO2 Sulphur dioxide
USEPA United States Environmental Protection Agency
VOC Volatile organic compound

14
6.0 References

ABARE (1999). Australian Energy: Market Developments and Projections to

2014-15, ABARE Research Report 99.4, Australian Bureau of Agricultural and
Resource Economics, Canberra.

Australian Gas Association (1999), Gas Statistics Australia, The Australian

Gas Association, Canberra,

CARB (1991). Identification of Volatile Organic Compound Species Profiles,

ARB Speciation Manual, 2nd ed, vol1, California Air Resources Board,
California, USA.

EA (1999). National Pollutant Inventory Emission Estimation Technique

Manual for Combustion in Boilers, Environment Australia, Kingston, ACT,
Australia.

USEPA (1992). VOC/PM Speciation Data System – Version 1.50, United

States Environment Protection Agency, Research Triangle Park, North
Carolina, USA.

USEPA (1996). Compilation of Air Pollution Emission Factors, vol. 1,

Stationary Point and Area Sources, AP-42, 5th edition, Supplement B,
Chapter 1.5: Liquified Petroleum Gas Combustion, USEPA, Research
Triangle Park, North Carolina, USA.

USEPA (1998). Compilation of Air Pollution Emission Factors, vol. 1,

Stationary Point and Area Sources, AP-42, 5th edition, Supplement D,
Chapter 1.4 Natural Gas Combustion, USEPA, Research Triangle Park, North
Carolina, USA.

15