Unned States EPA- 600/7- 91-002
Environmental Protection
*•"*» Januarv 1991
&EPA Research and
Development
APPROACH FOR ESTIMATING
GLOBAL LANDFILL
METHANE EMISSIONS
Prepared for
Office of Air and Radiation
Prepared by
Air and Energy Engineering Research
Laboratory
Research Triangle Park NC 27711
-------
EPA-600/7-91-002
January 1991
APPROACH FOR ESTIMATING
GLOBAL LANDFILL METHANE EMISSIONS
By.
Rebecca L Peer. Ann E. Leinlnger, Barbara B. Emmel, and Susan K. Lynch
Radian Corporation
3200 Progress Center
Post Office Box 13000
Research Triangle Park, North Carolina 27709
EPA Contract 68-02-4288
Work Assignment 48
Project Officer. Susan A. Thometoe
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, North Carolina 27711
Prepared tor.
U. S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
ABSTRA CT
in response to concerns about global warming. the U. S. Environmental
Protection Agency’s (EPA) Office of Research and Development (ORD) has
initiated a program to characterize the effects of global change, including
identifying and quantifying emission sources. EPA’s Air and Energy
Engineering Research Laboratory (AEERL) is part of this effort, and is
particularly concerned with quantifying emissions sources both in the
United States and globally.
This report provides an overview of the available country-S specific data
and modeling approaches for estimating global landfill methane. The
current estimates of global landfill methane indicate that landfills
account for between 4 and 15% of the global methane budget. The report
provides sri approach for using country-specific data and field test data to
develop a less uncertain estimate of global landfill methane. The
development of enhanced emissions factors for landfills and other major
sources of methane will improve the understanding of atmospheric
chemistry and feedback effects, will target mitigation opportunities. and
will ensure cost- effective mitigation strategies.
Li
-------
TABLE OF CONTENTS
Se ct ion
1 Introduction 1-I
2 Conclus ions 2 - 1
3 Review and Assessment of Landfill Models and Data Availability . 3- 1
4 Development of A Global Landfills Model 4-i
5 References 5 -1
APPENDICES
A Waste Composition A-I
B Solid Waste Management Methods In Eleven CouniTies B-i
C Sample Landfill Survey Form 0-1
U I
-------
FIGURES
Fioure Paae
ConceptS Scheme for a Global l.andfL Methane Model 4-2
2 Lettering and Numbering Conventions for Grid Cells 4-12
3 Indian Cell Landfill Methane Emissions. Quadrant A, 1° x f Gild 4-14
4 Indian Cell Landfill Methane Emissions. Quadrant 8,1’ x 1’ GrId 4 -15
5 European Cell Landfill Methane Emissions, Quadrant B, 1’ x 1’ Grid 4 .19
6 European Cell Landfill Methane Emissions, Quadrant C, 1’ x 1’ GrId 4-20
TABLES
Table Paae
1 Landfill Gas Generation Rates end Refuse Moisture Content 3-8
2 Effect of Moisture Content on Land! M Gas Generation Rate 3-9
3 Dam UsedforRegresslonof Waste GorseratlonRate on GNP 4-7
4 Cl’mractedstlcs of the Indian Cell 4—10
$ Characteristics of the European Cell 4—Il
6 Reat isofAflocatlonAna ly e lsforind lanCefl 4-13
7 Input Data for European Cell 4-16
8 Results of hilocatlon Analysis for European Cell 4-18
iv
-------
1.0 INTRODUCTION
In response to concerns about global warming, the U.S. Environmental Protection Agency’s (EPA)
Office of Research and Development (ORD) has Initiated a program to characterize the effects of global
change. Including Identifying and quantifying emission sources. EPA’s Air and Energy Engineering
Research Laboratory (AEERL) Is part of this effort, and Is particularly concerned with quantifying
emissions sources both In the United States and globally.
Considerable effort has beer. expended studying carbon dioxide (C0 2 ) emissions since CO 2 Is
responsible for most of the global warming. Methane (CHJ, Is of particular concern since its radiatrie
forcing potential has been estimated to be 20 to 30 tImes that of CO2 on a mole basis; furthermore,
atmospheric methane is increasing at a faster rate than any of the other greenhouse gases except for
CFCs (Rodhe, ¶990). Although the major sources of methane are known qualitatively, constderabte
uncertainty exists about the quantitative emissions from each source. One of the goals of AEERL’S
global climate research program is to develop better models and Inventories for methane sources.
This report summarizes the current state of knowledge with respect to one Important methane
source—landfills. The objectives of this study are:
• to evaluate the approaches cunenhly available for estimating landfill methane emissions.
and
• to determine the best available approach for estimating global methane emissions from
landfills.
These objectives were met by reinewing the current literature on methanogenesis in landfills, collecting
and evaluating methane emissions models, and IntaMewlng experts In this field.
The best approach Is obviously determined by a variety of factors Including the desired level of
accuracy, desired resolution, data limitations, and budget and time constraints. The level of accuracy Is
largely determined by the needs of the users of the model outputs. Policymakers need quantitative
measures of landfill emissions in order to develop mitigation strategies and to assign priorities to
mitigation programs. However, they may need only one number, such as average annual global
methane emissions from a given source; the finest resolution they may need is likely to be at the
country-specific level. At the other end of the spectrum, AEERL’s model may be needed to supply
irdomialion to regional and global atmospheric models. If so. the resolution of the data will need to be
finer. Spatally ,theemls&onsmaybeneededforgrldceilsaslaSl0’XlO’OraSSrnaIlaSfXf.
Temporally, time periods smaller than a year may be desirable.
1-1
-------
At this time. AEERL can only recognize that these dnrergent needs exist, but can not say for
certain that the needs of all users can be met. The limitations to meeting all these needs are partly
related to the costs of model development Even more critical, however, Is the large amount of
uncertainly associated with modeling methanogenesl& Cost considerations aside, the data requIred as
Inputs for a mechanistic model of methane production may not e dst
The conclusions and recommendations are summarized In SectIon 2.0. SectIon 3.0 discusses
several different modeling approaches that are currently available. and discusses data needs and
availability. Section 4.0 presents a conceptual scheme for a global landfills model, and outlines a
program to develop that model further.
1-2
-------
2.0 CONCLUSIONS
The purpose of this project was to determine the preferred approach for estimating global landfill
methane. This effort Is part of a larger effort In which field test data are being collected to develop
enhanced emission factors for major sources of methane Including coal mines, natural gas
production/distrIbution systems, and waste disposal facilities such as landfills. The development of
enhanced emission factors will Improve the understanding of atmospheric chemistry and feedback
effects, will target mitigation opportunities, and will ensure cost .effective mitigation strategies.
This report provides an overview of the approach to develop enhanced emission factors for global
landfill methane. The approach was developed considering the availability and quality of country-specific
information such as the amount of waste being landfilled, the composition of Iandfiled waste, and other
landfill characteristics that affect landfill methane emissions. The approach was also developed
considering the needs of policymakers and atmospheric chemistry modelers.
Provided below are conclusions from this project for developing an approach to more reliably
estimate global landfill methane:
1. An analysIs of available models was conducted (SectIon 3.1) and the conclusion
Is that several current models edst which could be modified for use with
country-specific data. The models were evaluated considering availability of
required Inputs. In future work, field test data are to be collected to evaluate
how the available models compare to gas production data at landfill sites where
methane is being collected arid controlled/utIlIzed. The results of this future
work wIll be used to define the algorithm for estImating global landfill methane.
2. A review of available country-specIfic data was conducted (Section 3.2). It was
found that data on waste composition and waste management for most
countries are adequate. However, some regions are not covered as well as
others, so extrapolation from similar countries or use of surrogates Is needed to
develop inputs for those countries. 4J 5 O Important are the country-specific waste
generation rates because methane production is directly proportional to the
amount of degradable waste Iandf Iliad. Date are available for a large number of
countries. it was also found that gross national product (GNP) Is a good
predictor of waste generation rate. Future work will need to determine whether
to use estimates of waste generation rates or to use the gross national product
approach developed by Keith Richards of the United Kingdom Department of
Energy.
3. The results of enhanced emission estimates for global landfill methane will be
used as Input to other models. Policymakers are generally concerned with total
and country-specific estimates. Atmospheric chemistry modelers generally
require finer resolutiorL A sensitivity analysis of three allocation schemes was
conducted to test the relatwe performance of each scheme using a range of grid
sues: 1’ X 1’,5 x5’,and 10’ xlO ’. The allocation schemes range fromthe
very simple to the very detailed. The conclusion of this analysis is that a
Population Cenlrold methodology previously developed for a global VOC
2-1
-------
Inventory study Is reasonably accurate and cost-effective for a 10 x 10 grid.
It maximum flexibility Is desirable or If a? X? grid cell Is required, then a new
methodology Is required (Section 4.1). Future work will need to determine the
user needs at enhanced emission estimates for global landfill methane.
4. A review was conducted of the functional relationships between methane
production and the various factors known or suspected to affect the rate of gas
production. it was found that moisture appears to be the greatest factor
affecting gas production. The other factors such as waste age, composition.
quantity and quality of nutrients, and ambient temperature may also affect gas
production. However, the results of studies reported In the literature are
sometimes contradictory and It Is unclear how the results would be extrapolated
to actual landfUis Future work where field testing date are to be collected will
help In determining the functional relationships for the factors affecting gas
production. In addition, work being conducted In the United Kingdom, Sweden.
the Netherlands, and india will also provide data needed to help determine the
functional relationships. The results of this work need to be collected for
developing the Inputs for estimating global landfill methane.
2-2
-------
3.0 REVIEW AND ASSESSMENT OF LANDFILL MODELS AND DATA AVAiLABILIT Y
Currently available models are reviewed below In Section 3.1. Data gaps and knowledge gaps
that limit development of a global model are Identified In Section 3.2.
3.1 EVALUATION OF AVAILABLE MODELS
Three different types of models dealing with landfill methane were Identified. The first are global
landfi l l methane emissions estimation models. These are simplistic models which do not take Into
accotfl time-dependent variations hi methane production. In general, the refuse generated annuallyjs
assumed to be converted to methane and carbon dioxide based on the degradable carbon content in
the same year that the waste is landfilled. Spatial variation is, at best, limited to the scale of individual
countries.
The second group of models are theoretical first-order kinetic models of methane gas production.
They are based on the methanogenic processes of bacterial populations and include a time component.
These models are generally applied to individual landfills, although they could be applied to entire
countries or regions.
The third group of models are not concerned with methane production, but model the movement
of gases through a landfill. While these models are capable of providing the most detaIled emissions
data with respect to temporal resolution, they are far too detailed for global applications. These models
ate not discussed any further In this report The first two types of models are described in more detail
be .
3.1.1 9lobal Emissions Models
Two examples of global models were found. The simplest was developed by R i chards (1989) and
Is based on using Gross Domestic Product (GDPJ.’ Using estimates of annual refuse production and
GDP for the U.S. and western Europe, RIchards calculated that these two Industrialized regions (with
account for 65% of the world GDP) produce 492 million metric tons of refuse per year. Making some
other gross assumptions about gas generated per ton of refuse arg f percent Iandfliled, he estimates that.
globally, 392 x 10 m 3 landfill gas are produced each year. This number Includes all landfill gases, so
the amount of methane was extracted using an assumed C0 3 /Ctt ratio. Richards estimates that 9.8.
18.3 million metric tons of methane am emitted from uncontrolled landfills globally.
‘Gross Domestic Product Is the Gross National Product excluding payments on foreign Investments.
3 - 1
-------
Another example of a method for estimating global landfill emissions Is provided by Bingemer and
Crutzen (1987). In this methodology, 80 percent of the degradable organic carbon (DOC) In landfill
waste Is assumed to be converted Into Landfill gas that is 50 percent methane by volume. The authors
then used existing studies to develop estimates of annual refuse generation rates and waste composition
for different countries. Data are not available for all countries, so estimates had to be made for the
USSR and eastern Europe and for some developing countries. Based on these assumptions the authors
estimate global landfill emissions of 30-70 million metric tons annually.
These two methods produce quite different estimates, reflecting the problem with an approach
that uses such gross simpllficatlons. Bingemer and Crutzen (1987) assume a rather high conversion rate
(80%) which Is not hke ly to apply to all climates and landfill types. They also assume a relatively high
percentage of waste Is Iandfihled (80% globally). In fact, the percentage of the waste deposited in
sanitary landfills where anaerobic conditions are likely to occur may be much lower than 80 percent. In
developed countries, alternative forms of waste management, such as Incineration, recycling, and
composting, are becoming increasingly impovtant (Richards, 1989; Swartz, 1989). In the developing and
undeveloped countries, very little waste Is deposited In sanitary landfills. Most of ft ends up in dumps
(i.e., deposited on the suiface) where aerobic conditions prevail (Bhlde et a!., 1990).
Although consideration of all these variables suggests that Bingemer and Cnitzen’s estimates are
on the high side. Richards’ estimates are based on a few broad assumptions and must be regarded as
crude appro,dmatiorzs. Both of these approaches have merit and the techniques used may be
Incorporated Into more detailed models. However, at this time, both methods yield only rough
approxirnatlorts of presentday emissions; they have even less credibility for projecting emissions In the
future.
3.12 Methane Gas Production Models
Methane emissions from IridMdual aridfilis may be estimated using theoretical first-order kinetic
models of methane production. Specific models are discussed and evaluated in Emcon (1982). These
and other models were reviewed for EPA ’s Office of Air Quality Planning and Standards (OAOPS) by
-------
Radian CorporationP The Scholl Canyon model has been modified for use in estimating landfill VOC
emissions in the United States.
The original form of the Scholl Canyon model Is
°C 1 1 4 L R(e t -e )
where:
00114 = methane generation rate at lime t, ft 3 fyr
= potential methane generation capacity of the refuse, f9/Mg refuse
P = average annual refuse acceptance rate during active life of landfill, Mg/yr
k = methane generation rate constant, 1/yr
c time since landfill closure, year (c = 0 for an active landfill)
t time since the initial refuse placement, year
For a given landfill, R, c, and t are usually available. Values for l. and k are not so easy to find, in part
because they are defined ambiguously. The potential methane generation capacity of refuse, l.a, S
generally treated as a function of the moisture content and organic content of the refuse. The rate
constfl Ii, Is a function of many factors, Including moisture, pH, temperature, and other environmental
factors, as well as landfill operating characteristics.
Unfortunately, no explicIt functional relationships are available that can be used to estimate the
kinetics of methane production. Both L , and k must be estimated; the OAOPS method Is to use
measured methane emissions from several landfills to calculate both k and L 0 for different climates within
the United States. Mother approach Is to estimate L 0 for a given landfill from refuse composItion; then,
using measured methane generation rates, k can be calculated for that landfill.
2 Memorandum to S.& Thomeloe, EP& from Y.C. McGuinn, Radian Corporation. “Use of a LandfIll Gas
Generation Model to Estimate VOC Emissions from Landflls. June 21, 1988. ll-B.14, U.S. EPA
Standards of Performance for New Stationary Sources and Guidelines for Control of Eidstkig Sources:
Municipal Solid Waste Landfill Docket, Docket No. A-88-Ct
3 Memoranclum to S.& Thomeloe, EP& from Y.C. McGuinn, Radian Corporation. “Sensitivity Analysis of
Landfill Gas Generation Model? June 21, 1988. Il-B-iS, U.S. EPA Standards of Performance for New
Stationary Sources and Guidelines for Control of Existing Sources: Municipal Solid Waste Landfill
Docket, Docket No. A-8&-Ot
3.3
-------
Similar models are described in Emcon Associates (1982). They will not be discussed here since
the limitations of the Scholl Canyon model for global modeling apply to all of these kinetic models. A
sensitivity analysis of the Scholl Canyon Mode!’ found that emission rate was a direct linear function of
L 0 and R. and a negative exponential function at k. The sensItivity 01 emIssion rate to k depends partly
on the year (t) arid partly on the magnitude of I c. Furthermore, greater uncertainty Is associated with It
in the U.S.. k appears to be affected somewhat by climate li i that estimated k values tended to be higher
In states with higher precipitation. However, a wide range of k values were found witlin both dry’ and
wet’ states.
These models would have to be modified for use on a global scaJe for many reasons. One
problem Is that the model requires Informat ion on landfill age and time since closure. It would be
ImpossIble to gee this kind of Information for every landfill within a country. An alternative would be to
model the average landfill within each country and multiply model emissions by the number of landfills.
However, this still requires fairly detailed country•specific data.
An even more important problem Is that k must be estimated from empirical data on methane
generation rates. While this Is a reasonable approach for the United States. It is probably not feasible on
a global scale because sufficient data on methane generation Tales are not likely to be available to
estimate k values for Individual countries or regions. If functional relationships between Ic and the factors
believed to affect k (e.g.. moisture) were available, kinetic models such as these might have more
potential. Even I I these relationships could be determined, unless a rather large data set covering a wide
range of refuse types (for estimating L 0 ) and moisture regimes (assumed to partially determine Ic) Is
available, this approach Is unlikely to be any more reliable or accurate than the methods discussed In
Section 3.1.1.
32 DATA NEEDS AND AVAJLAEIJ Y
Data necessary to develop global models for methane emissions from landfills Include refuse
generation, waste composition. landfill size, aerobic vs. anaerobic processes, age of refuse. pH.
temperature, moisture content, arid available nitrients. Researcher’s opinions vary concerning which
parameters are most Important In determining methane emissions from the landfilL For eample, while
many studies conclude that a certain pH range combined with moisture and refuse content will result In
maximum methane generation, one study found that lack of soil noleture actually Increased methane
generation from the landfill (Jones and Nedwell, 1990). Conclusive data concerning these parameters
‘See footnote 3.
s - s t
-------
are important in developing modeling approaches. Data may be collected by initiating sampling of
landfills or bench scale research.
Availability of data concerning landfill methane emissions was assessed through database and
library searches using keywords such as: landfill, municipal solid waste. biomass, alternative energy.
methane, and anaerobic decomposition. information was evaluated to establish relevance to this
project. The results of these searches reveal that there is a great deal of Information concerning waste
composition and, to a lesser extent, waste generation rates. Information concerning factors that affect
the generation of methane from landfills Is generally not available, or. where available, Is not conclusive.
information concerning waste composition and generation as well as disposal methods was found
for both developed and developing countries. Composting is becoming a popular program for reducing
the amount of waste sent to landfills in some countries. Tables summarizing this information can be
found in Appendices A and B.
3.2.1 Essential Data Requirements
In order to develop a modeling approach to estimate global landfill methane emissions, Inputs for
parameters affecting methane generation are necessary to develop estimates. The major data types
required are discussed below. The first two, waste composition and refuse generatIon rate, affect
potential methane production. The erMronmental variables affect the rate of prOductiOrL
3.2.1.1 Waste Composition and Refuse Generation Rate-
Waste composItion and refuse generation Is a major determinant of both gas quality and rate of
production. Both organic content of the waste as well as the sIze of the particles of waste influence gas
generation quantity. Waste having a high percentage of biodegradable organic material (food and
garden waste, paper, wood) and small particle sIze (25-250 mm) has been found to Increase gas
production from landfills, although other factors, such as the pH, affect the concentration of methane as
opposed to carbon dioxide (SenIor, 1990).
Refuse generation rates affect the amount of waste being delivered to a landfill and therefore the
amount of waste available for decomposition. Waste decays at different rates. According to one source
(Rovers at aL 1977). food and garden waste decompose within 1.5 years, paper breaks down In
5-20 years. arid wood may take 20-100 years to decay. Refuse generation rates In past years will affect
the amount of material degrading in the landfill at any given time. Refuse generation rates can also be
3.5
-------
used to estimate the amount of biodegradable carbon In a landfill given annual generation rate,
composition of waste, and size of the landfill.
Industrial ized countries tend to have high generation rates, with paper end cardboard as the
most significant contributors to degradable carbon. Waste In developing countries, by contrast Is
generally comprised of vegetable arid garden refuse, but the total volume of waste generated Is lower.
Some data on waste generation rates In different countries are also available (e.g.. World Resources
InstItute, 1988). See Appendix A for a summary of refuse composition in developed and developing
countries.
3.2.1.2 EnvIronmental Variables—
Several environmental parameters influence methane production from landfills: pH. moisture
content, refuse generation, waste composition, temperature, aerobic vs. anaerobic processes, landfill
size and type. Data are needed to quantify relationships between these variables and how they affect
the methane generation rate. Much of the recent research on landfills has focused on optimizing
methane production for collection and utilization as an energy source.
Moisture-in landfills, moisture determines the mixing. dilution, and flushing of refuse components
and nutrients. Moisture has been found to Increase methane production k landfills; however, In some
Instances, decreased moisture values have increased methane production, possibly because ft
decreases the activity of methanotrophlc sail microorganisms which consume methane (Jones and
Nedwell, 1990).
ft Is generally accepted that controlling water conditions can Increase methane generation rates;
however, Introducing controls may also Introduce problems (SenIor, 1990). Presence of high water
content should enhance availability of nutrients and, therefore, stimulate bacterial growth. However.
there Is a distinction between moisture volume and Infiltration. One study performed with three cells
fIlled with different amount of refuse found that rapid Infiltration co d Impede methane generation
(R iers and Farquhar. 1973). Another study ( IQink and Ham, 1982) found that methane production
could increase 25-50 percent with Infiltration even when total moisture content of the refuse remained
constant. This Increased production rate of methane may be due to an Increase In uniform distribution
of nutrients and pH QQink and Ham, 1982).
No definitive answer has been found concerning the necessary moisture content for maximum
methane generation. Laboratory studies (deWalle et al. , 1978) have found that maximum rates result
from water-saturated refuse; while other studies found that moisture contents of 60-80 percent produce
-------
maximum methane generation (Farquhar and Rovers, 1973). Researchers hypothesize that large water
additions may Introduce oxygen which delays the initiation of methanogenesis, and In some instances,
may Introduce acldogenesis which retards methane production. Conversely, the ratio between methane
and carbon dioxide may Increase In favor of methane production, in the presence Of elevated moisture
contents (SenIor, 1990).
In a study conducted by EPA ’s Office of Solid Waste (05W), landfill gas generation rates between
0.34 and 15.28 cubic meters of landfill gas per cubic year of refuse per year are reported. These data
are provided In Table 1 (SCS Engineers, 1986). Assuming a refuse densIty of 59324 kIlograms per
cubic meter, this corresponds to a landfill gas generation rate of 0.75 to 34 Alters of landfill gas per.
kilogram of refuse annually. One Important finding of this study was the correlation between landfill gas
generation rate and moisture content. Based on these data obtained from 12 landfills in °wet° States and
8 landfills In Southern Califomia. emissions from wer landfills are approximately 2.6 times greater for the
wer States as for the dry° ones. The field data supporting this factor is presented In Table 2. The
factor of 2.6 is obtained when the mean or median value (7.78 or 7.67) of wet region gas generation rate
is divided by the dry region gas generation rate (3.04 or 3.00). The ‘wer States are defined as the
States wIth annual precipitation of 23 inches (58.4 centimeters) of annual precipitation. All Sates except
the following receive greater than 23 inches of precipitat ion annt Wy: Arizona. California, Co lo rado,
Hawaii, Idaho, Montana. Nevada, New Mexico, North Dakota, South Dakota, Utah, and Wyoming.
I nnj -M optimum temperature range exists In which the inethanogenlc bacteria function
best ft Is the temperature of the anaerobic zone that regulates the optimum methane production.
Research has found that at 35°C almost 80 percent of the degradable organic carbon (DOC) may be
disslmllated. ft Is assumed that 80 percent of the DOG Is convened Into blogas contaInIng 50 percent
by volume of methane (Bingemer and Cnnen. 19871. Verstraete et i i. (1984) obtained a 70 percent
Increase In gas production with a temperature elevation from 22 to 330 C.
Temperature in the landfill is determined by microbial metabolism, dry density of refuse, specific
surface area, refuse composition, and water content. Landfill temperature changes In response to air
temperature changes have been reported (Rovers arid Farguhar, 1972). For example, a landfill In
Canada exhibIted seasonal temperature fluctuations between 2 and 21°C at a depth of 1.22 in.
Temperature can also affect gas composition. Temperature increases can affect the fermentation
balance, resulting In Increased acid generation while inhibiting methanogenesis (KasalL 1986).
3-7
-------
TABLE 1. LANDFILL GAS GENERATION RATES AND REFUSE MOISTURE CONTENT
Gas
Refuse
Generation
Moisture
Methane
Rate
Content
Content
Landfill Location (rna/rt 3 yr)a
(WI %)
(vol %)
WET REGIONS
Michigan 2.67 33 52
Maryland 3.41 34 49
WisconsIn 5.33 50 52
New York State 7.04 53 57
Washington, DC 4j 7.18 — 47
Maryland 7.37 - 56
Florida 8.15 29 55
Ohio 8.29 33 53
Florida 9.26 24 50
New York State 9.26 30 51
Ohio 9.59 37 55
Florida 20.00 42 58
DRY REGIONS
Southern CalifornIa 0.44 17 56
Southern CalifornIa 1.78 12 52
Southern California 222 18 55
Southern California 2.30 16 50
Southern California 3.67 27 54
Southern CalifornIa 3.92 18 56
Southern California 4.6? 22 54
Southern California 5.30 22 51
Cubic meters of landfill gas/cubic meters of refuse per year.
-------
TABLE 2. EFFECT OF MOISTURE CONTENT ON LANDFILL GAS GENERATION RATE
Gas
Refuse
Generation
Moisture
Methane
Rate
Content
Content
Landfill Location (m /rn 3 -yr)
(wt %)
(vol %)
ET REG!ONS
Michigan 2.67 33 52
Maryland 3.41 34 49
Wisconsin 5.33 50 52
New York State
Washington, DC Area 7.18 47
Mar ,1ancI 7.37 — 56
Florida 8.15 29 55
Ohio 8.29 33 53
Florida 9.26 24 50
New York State 9.26 30 51
Ohio 9.59 37 55
Florida 20.00
Mean 7.78 37 53
Median 7.76 34 53
Standard Deviation 4.52 9 3
DRY REGIONS
Southern CalifornIa 0.44 17 56
Southern California 1.78 12 52
Southern California 2.22 18 55
Southern CalIfornia 2.30 16 50
Southern CalIfornia 3.67 27 54
Southern California 3.92 18 56
Southern California 4.67 22 54
Southern California .
Mean 3.04 19 54
Median 3.00 18 54
Standard Deviation 1.63 5 2
‘Cubic meters of landfill gas/cubIc meters of refuse per year.
3-9
-------
ps—The pH value in a landfill can vary at different depths and ceils of the landfill. Ranges of pH
have been reported as 4.4-6.9 (Bookter and i-tarn, 1982); however, the optimum for inethanogenesls
seems to be around 7.0 (Emcon Associates. 1982). Low pH values and high concentrations of
carboxylic acids can kihibit methanogenesis. While the methanogenlc bacteria necessary for production
of methane are most active at certain pH valuet this Is a parameter that would be difficuft to measure In
the field as It is highly variable In space and time.
Aerobic vs. Anaerobic Conditions—During aerobic processes In the landfill, bacteria decompose
refuse by consuming oxygen while producing carbon dioxide and water. Anaerobic processes result In
the production of methane. The aerobic decomposition of waste generally lasts only a few weeks while
the anaerobic process can continue for 10-30 years after the filling has been completed (Bogardus.
1987). Some methane produced through anaerobic processes in the landIl i l may be oxidized by
meth)lotrophic bacteria present In the top cover of the landfill (Jones and Nedwell, 1990).
Waste disposal practices dictate the presence of aerobic vs. anaerobic processes Landfirllng
wastes encourages the presence of anaerobic processes while composting and surface disposal.
prevalent in developing countries, result In aerobic processes of decomposition.
Size and T ime of Landfill-Landfills may Include municipal solid waste. Industrial waste, hazardous
waste, or In some cases, a combination of rifuse. Because It Is Important to have waste with a high
percentage of organic material for methane production, municipal solid waste landfills are the obvious
source of methane emissions. Hazardous and Industrial waste landfills n .y contain compounds that will
restdt In a low pH atmosphere tSc to the methanogenlc bacteria.
Larger landfills will generally provide greater n asa of orgSc materiaL However, no information
was found to describe the functional relationship between landfill size arid methane production. The
depth and surface area are probably more knportant factors to consider than size alone. For example, a
large shallow landfill Is likely to produce more CO 2 than a deep landfill of comparable size because of Is
greater surface area.
-------
4.0 DEVELOPMENT OF A GLOBAL LANDFILLS MODEL
Ideally, the global landfill methane emissions model used by AEERL should meet two criteria. The
first is that It provides reliable country- or region-specific estimates of methane currently produced by
landfills. The second Is that It be capable of projecting emissions into the future. The rehability of these
projections will depend on the ease with which the models parameters can be changed to reflect
various world scenarios; for example, country-specific trends in waste management could affect the
amount of waste IancJfihled as opposed to Incinerated or composled. Increasing affluence of developing
countries may Increase the amount of waste generated annually. Finally, changes In the earth’s climate—
particularly In precipitation—could affect the rate of methane production.
4.1 CONCEPTUAL SCHEME FOR A GLOBAL LANDFILLS METHANE EMISSIONS MODEL
Based on the review of the currently available models, the needs of AEERL and other model
users, and the current state of understanding of methane production, a general scheme for a global
landfills model can be described. This scheme Is presented solely for the purpose of synthesizing the
current state of the science. ft provides a framework for discussion, and helps identify both data needs
and potential modeling methodology requirements. The actual form of the model that Is developed may
be quite different.
Figure 1 shows a flow diagram of the conceptual model. Three modules are delineated that
reflect the two major steps required to generate annual methane emissions per country plus a third step
that allocates those emissions to a global output format A brief summary of each step Is:
(1) DetermIne the methane potential of landfill waste. taldng Into account refuse composition,
level of development of the country, and other pertinent factors:
(2) Calculate methane generation rate on an annual basis, taking Into account the factors that
affect the rate of production, annual waste generalion rate, arid methane potential of the
waste; and
(3) Allocate methane emissions to a spatial grid for Input Into other models and data bases
(such as atmospheric models or mapping programs).
Note that only steps I and 2 are required to get country-specific and total global methane emissions.
Thethieplsnecessa ryovdyttthedataaretobeusefultoatmosphedcmodelersorifaspalia ldata
base is used for the inventory. These steps are discussed In more detail below.
4.1.1 Calculatino Methane Potential
The organic content of the waste primarily determines the landfill gas potential of the waste.
However, the relatwe amount of methane In that gas is determined by the disposal method. If wastes
are incinerated or composted, only CO 2 will be produced. Wastes that are dumped’ (i.e., surface
disposal In relatively shallow heaps) will decompose primarily under aerobic conditions, producing CO 2 .
In addition, these trash heaps are open to scavenging animals that will remove a good portion of the
4 _I
-------
A
Agure 1 . Conceptual Scheme for a Global Landfills Methane Model.
du gnI
4-2
-------
biodegradable wastes. Only wastes that are deposited in sanitary landfills (i.e.. excavated pits that are
tilled and covered with some sort of cap) are likely to produce significant amounts of methane.
In Industrial countries, It is probably safe to assume that landfill wastes are going to sanitary
landfills where anaerobic conditions prevaIl. The relative percentage of potential methane will depend on
the organic content of the refuse, and the amount of waste that Is anaerobically gasified.
Various methods have been proposed for estimating mBthane potential. Theoretical estimates
range from 47 to 270 L CH 4 /kg wet composite refuse, although 31-94 L CH. 1 /kg wet composite refuse is
thought to be a more realistic range for most sanitary landfills (Emcon Associates. 1982). Relative
amounts of landfill gas constituents are generally predicted to be 54 percent methane and 46 percent
carbon dioxide. In fact, landfill gas Is typically 50-70 percent methane and 3D-5D percent carbon dioxide
with traces of other gases (Emcon Associates, 1982).
Module A of Figure 1 diagrams a series of categorizations the model would have to make to
determine methane potential of the waste for a given country. The fraction of biodegradable material
must be supplied as input data. The percentage of material that will be converted to Cu 4 Is determined
by a set of criteria. If the country is an industrialized one, then waste that is g.t incinerated, composted,
or recycled Is assumed to go to sanitary landfills. In this case, the relative methane content of the
landfill gas Is likely to be around 50 percent If the country Is not a developed one, then sanitary landfills
are lIkely to be found only in large cities. In the rural areas, open dumping is likely to occur. Moreover,
refuse composition may vary between rural and urban areas; In India, the organic content of refuse In
rural areas Is lower than In large urban areas (Bhlde at aL, 1990). Therefore, for these countries, two
estimates of methane potential may be needed: one estimate for urban wastes and one for rural wastes.
4.12 Calculating Annual Methane Emissions
Module B requires the following country-specific Input data: average amount of waste Iaridfilled
each year, the methane potential of that refuse (generated by Module A), and the methane generation
rate, k. The exact form of the methane generation algorithm is not specified. It could resemble one of
the methods described In Section 3.1, or it could be entirely different. A hybrid approach that uses
elements of Bingemer and Crutzen’s methodology and of the kinetics model would be to model annual
methane production within a country as though It were all being generated by one landfill. Waste
generated each year can vary as can k, the rate of methane production.
This approach does not require variables such as the age of the landfill. For estimating emissions
from an lndMdual sIte, the age Is very Important. However, for global emissions calculations, thIs
Information Is probably not Important A global model deals with a population of landfills; therefore, for
estimating current emissions, the population average Is sufficient, if a steady-state can be assumed.
then using current average waste production to calculate current emissions should give a reasonable
estimate (even though today’s emissions probably come from waste deposIted years ago). However, If
4-3
-------
the number of new landfills is Increasing at a rapid rate, then the average age of the landfills is
decreasing. Since landfill emissions tend to lag behind actual deposition of the waste, younger landfills
produce gas at a slower rate than older ones.
In fact, landililing Is likely to decline In developed countries as space becomes limiting and
incineration becomes more popular. On the other hand, sanitary landfills may become more common In
developIng countries, especially around cities. These two trends may cancel each other out. This Issue
needs to be recognized, but Is probably not the most Important Issue In the short-term (!.e., the next
20-30 years). ft Is more Important for doing projections into the future, and can probably be dealt with In
the model by use of a lag time. Other variables, such as refuse acceptance rate and time since closure,
are easily dealt with at this scale. Refuse acceptance rate Is simply all of the waste sent to landfills each
year. That amount can be allowed to change by modeling it as a function of population or economic
growth. Time since closure Is no longer relevant, unless changes in waste management practices are
foreseen such that all landfills within a country are to be closed.
One critical piece of information is still needed: the functional relationship between k and the
suite of factors that determine k. As was stated previously, a variety of factors such as moisture affect k,
but the functional relationships are unknown. Globally, landfills may be found In a wide range of
climates. Since both moisture and, to a lesser extent temperature, are known to affect methane
production rates, developing a better understanding of the functions! relationship between k and these
two variables will increase the reliability of methane emissions estimates In several ways. First, for
modeling current emissions, geographic variability in emissions will be more accurate. ft may even be
possible to Increase the temporal resolution to a seasonal scale for those countries with pronounced wet
arid dry seasons. Second, the model can be used to do projections Into the future for alternative
dimate scenarios. The possibility of feedback effects where climate changes accelerate the release of
RrTGs Is one that concerns scientists and pollcymakers. To the extent that landfill methane production
is affected by climate, feedback effects are potentially Important.
Unfortunately, as was discussed in Section 32.12, InsufficIent data are available to develop the
needed functional relationships. An Interishie research effort Is needed to gather data that can be used
to develop this component of the model. SectIon 4.2.3 discusses the proposed research In more detail.
4.1.3 Allocation to a Grid
The results of the emissions model will be needed as Input to other models. partlculady
atmospheric models. These models are typically spatial with resolutions from 10 x 1’ up to 10’ x 10’. In
order to be compatible with these models, the methane emissions model will need to convert the
country-specific emissions output to emissions per grid cell. This poses several problems: how wIll
emissions be allocated to the spatial grid, how small will the grid cells need to be, and what effect will
choice of grid cell size have on the validity of the allocation scheme?
-------
At this time, the desired grid cell size is unknown. In fact. It is likely that methane emission model
estimates will be needed for Input into several different models with different resolutions. Most general
cIrculation models (GCMs) are relatwely coarse with resolutions of 10° x 10’ or 5° x 5° Regional models
may require data at a finer resolution, such as 1° x 1’. Clearly, developing an allocation scheme for a
1’ x 1° grid will be extremely labor Intensive, and should not be done unless really needed. On the other
hand, if a 101 x 10° grid Is used Initially, Is there some way to design an allocation scheme that can be
adapted to a finer resolution If the need should arise in the future? In other words, can a flexible
allocation scheme be devised that allows for different grid resolutions, but is also not too costly to
develop?
If the exact location of every landfill In the world were known, then an allocation scheme would
not be needed. However, the acquisition of Information this detailed Is not feasible. Several possible
schemes can be envisioned. The simplest would be to allocate emissions uniformly within a country; the
relative area of the country Included within a particular cell would be used to weight the emissions from
that country assigned to that cell. For example, if 25 percent of a country is located within a given grid
cell, then 25 percent of that country’s emissions are allocated to that cell.
Another approach is to use population centroids. The problem with this approach is that the
number of centrolds used Is often dependent on the grid cell size. This method was used in a global
voc lnventor9 which allocated VOC emissions to sb’ x 10° grid. Population centroids were
developed for each cell, but the number of centrolds varied. Every cell had at least one centroid unless
there were obviously no emission sources within the cell (e.g.. a cell that had only open ocean). Over
populated areas, the number of centrolds depended less on the size of the population than on Its
distribution geographically. For example, a cell with a large population that was concentrated in one
corner of the grid would have fewer centrolds than one with a smaller but more evenly dispersed
population.
Many other possible schemes exist The problem for this particular model Is In trying to identity a
scheme that will be flexible but also not too costly to Implement.
42 FIWNG THE GAPS
The review of models and data In Section 3.0 end the discussion of a conceptual model above
have IdentIfied strengths and weaknesses of currently available models and data. The remainder of this
report addresses remedies for three of the issues Identified:
(1) the use of economIc Indicators to predict refuse generation rates (as In Richards’ model);
(2) the relative merits of emission allocation schemes as a function of grid cell size; and
5 Details on the development of this Inventory will be available in a forthcoming EPA report.
4.5
-------
(3) the acquisitton of data that will allow development of a predictive methane production
model.
Each of these are addressed in more detail below.
Some other areas considered In more detail at this time are still important Refuse
composition by country Is an Important Input variable, but one for which considerable data exist (see
Appendix A). Although data are not available for many countries, reasonable extrapolations from
existing data can be made. Another Issue Is that of determining the proportions of waste going to
different types of treatments. Although some data have been found (Appendix B), It covers a small
number c i countries. little Is know about developing countries In particular. Furthermore, future trends
In waste management will make current data obsolete. Since the development data for individual
countries Is beyond the scope of this program, currently avanable sources will have to be relied on. For
most countries, the percent of waste Iandfii led will have to be estimated based on information for similar
countries.
4.2.1 Predicting Waste Generation Rates frorri GNP
Richards’ (1989) used gross domestic product (GDP) to predict waste generation rates In his
global estimation of landfill gas production. The basic concept Is that increasing affluence will be
accompanied by Increasing waste production. Since economic Indicators such as gross national
product (GNP) are widely available for most countries, this could provide a readily available surrogate for
waste generation rates.
To test this Idea, a smell sample (12) of countries representing a broad range of economic levels
was selected from a data set of waste generation rates (World Resources instItute, 1988). GNPs for
those countries were also given In the same source. The data are shown In Table 3; waste generation
rates and GNPs are not necessarily from the same years, but all are from the time period between 1980
and 1986.
A linear regression was used to test the relationship between waste per year and GNP. The
regression was significant with R 2 0.97. The regression model Is.
WASTE(1000 Mg yr) - 0.0399 • GNP (million US $). (1)
This looks like an excellent model even with such a small sample size. As a further test, the GNP of
India was used to estimate that cauntry s annual waste generation; this was converted to a per capita
A
-------
TABLE 3. DATA USED FOR REGRESSION OF WASTE GENERATION RATE ON GNP
Country
GNP
(million US $)
Waste Generation Rate
(1000 Mg/yr)
Canada
361.720
16,000
Costa Rica
3,790
634
Federal Republic of Germany
735,940
27,544
France
595,180
14,000
Ireland
18,190
1,270
Israel
26,730
1.400
Japan
1,559,720
41,095
Korea
98.370
15,746
Singapore
19,160
1498
Spain
188,030
10,600
United Kingdom
504.850
16,398
United States
4,221,750
178,000
4.7
-------
per day basis arid compared to an independent estimate ° The model prediction fell In the middle of
the range of the other estimate (0.2 to 0.5 kg/cap Ita/day).
However, both the U.S. GNP and waste generation rate are so much higher than the second
lamest (Japan) that this data point appears to exert undue influence on the model. The regression was
rerun without the U.S. data. Although the regression was still significant, the R 2 dropped to 0.88. The
regression model without the U.S. data Is
WASTE(1000JM yr) 3902 + 0.0251 • GNP (million US $). (2)
The intercept term In this model was not significant although It was very close (0.06). AgaIn, the model’s
prediction for India fell within the range of the Independent estimate.
This approach appears to be very promising but further research is needed. The best strategy
may be a mixed one, using different equations for different countries. The next phase of analysis should
Inc lude:
(1) Increased sample stze (more data observations are availSe);
(2) Separate analyses (or different regions (e.g.. continents); arid
(3) Separate analyses for industrIalized versus developing countries.
Other economic Indicators should be considered, such as per capita Income. However, these Indicators
need to be readily avai lable for a large number of countries If this method Is to be useful.
422 Allocation of EmissIons to a Grid
A sensitivity analysis of three allocation schemes was conducted to test the relative performance
ofeachschemeusigamngedgrldsiaes fx?,S’x5’,and lO”x lcr. Tlnllocallonschemes
range from the very simple to the very detailed.
The simplest scheme, UnlforzTC assumes emissions are uniformly ci istnibuted throughout a
country. Within a grid cell, emissions are calculated by determining what proportion of the country Is
Included In that cell, muItJplØng that proportion by total emissions for the country, and summing tar all
countries within the cell
The Population Centrold scheme uses an existing data base that was developed for EPA’s
Global YOC Inventory The population within a country Is subdMded into several groups based on the
5 Pe s communication from ‘tD. Bhide, National Erwironmental Engineering Research Institute,
Nagpur20, India, to D.L Campbell, Radian Corporation. Apr Il 13, 1990.
t 0
-------
The advantage of using the existing centroid data base is that It could be implemented at very
little cost, since the bulk of the development work has been done. The population data needs to be
updated but this will not require a great deal of effort. One potential problem wIth this scheme is that it
was developed for a 10’ x 10’ grId. The analysis performed In this study evaluates the performance of
the data set for finer resolutions.
The third scheme, ‘Urban/Rural.’ Is a composite of the first two. The population within a country
is divided Into two groups. The urban population is the proportion of people residing in cities of 200,000
or more. The rest of the population Is considered rural. Each city of 200,000 or more Is treated as a
population centroid. The populations of the centroids were summed and the total was subtracted from
the country’s total population. The remaining population was assumed to be uniformly distributed
throughout the country.
The Urban/Rural scheme was considered to be the most accurate of the three schemes used in
this analysis. Its outputs are used to evaluate the relative accuracy of the other two schemes. The
relative merits of all three schemes with respect to ease of implementation, flexibility, and performance
with differing resolutions were analyzed qualitatively.
Two 10’ x 10’ grid cells were chosen for analysis. The India cell lies between 70” and 80°E
longitude and 20’ and 30” N latitude. Part of Pakistan Is also Included In the cell. This cell was chosen
because It represents two cell ‘types’: It Is simple (only two countries included), and It Includes only
developing nations. The second cell was chosen to represent the other e remos: It Includes eight
countries, all of them Industrialized. This European cell Is bounded by 0’ to WE longitude and
40’ to 50° N latitude. Characteristics of each cell are summarized In TabLes 4 and 5. Each cell was
subdMded Into four 5° x 5° grids; one of these cells was further subdivided Into f x f grids. Figure 2
shows the lettering and numbering conventions used to identify each cell.
For the India cell, a waste generation rate of 0.125 Mg waste/yr/capita was used for India, and
0.017 for Pakistan. These waste generation rates were estimated using GNP values from World
Resources institute (1958) In the equation given In SectIon 42.1. Methane was assumed to be produced
at the rate of 30 rn 3 CK 4 /Mg waste. A second analysis for the Urban/Rural scheme only was run using
30 m 3 CH 4 /Mg waste for urban populations and 10 m 3 CI /Mg waste for rural populations (designated
by (b) In Table 6). ThIS is done to reflect the fact that sanitary landfills are relatively rare in developing
countries, and are only found In large urban areas. The methane potential used for rural area was
chosen arbitrarily as no estimates were found In the literature. The results of each allocation scheme for
10” x 10’ and 50 x 5° grids are shown In Table 6; results for two sets of the 1’ x 1° grids are shown In
FIgures 3 and 4.
In the European cell, the proportion of waste Iandfilled In each country was Included In the
analysis. Table 7 shows the waste generation rates and proportion landfilled for each country. A
methane generation rate of 30 m 3 Ct /Mg waste was used for all countries. The results for 10’ x 100
4-9
-------
TABLE 4. CHARACTERISTICS OF THE INDIAN CELL
Country
Total
Population
Proportion of
Country In
100 x 10” cell
IndIa
683,810,051
0.33
Pakistan
83,782,000
0.10
-------
TABLE 5. CHARACTERISTICS OF THE EUROPEAN CELL
Country
Total
Population’
Proportion of
Country In
10 x 10 cefl
Austria
7,507,000
0.02
Belgium
9,855,110
0.14
Federal Republic of Germany
61.658.000
0.25
France and Corsica
54,077.842
0.80
Italy and Sardinia
58,613,800
0.25
Luxembourg
364.000
1.00
Spain
37,430.000
0.06
Switzerland and Liechtenstein
6,391,180
1.00
4 -11
-------
r
V
0
>10
0
5 < ,
1
3
4
5
A
6
7
8
9
10
11
12
131415
16
17
18
1920
21
22232425
C
D
I
Agiare 2. Lenerthg end Numbering Conventions for Grid Cells
4.12
-------
TABLE 6. RESULTS OF ALLOCATION ANALYSiS FOR INDiAN CELL
Allocation Scheme
Emissions from
10°x lO °ceU
Emissions from
Sxrcel)
(10 m 3 CH 4 /yr)
(i
m 3 CH 4 /yr)
Percent of
100 x 10°
UnIform
850,488
A
B
C
D
192.210
234,735
188,808
234,735
22.6
27.6
222
27.6
Centroid
621,838
A
B
C
D
6,409
384,643
230.786
0
1.0
61.9
37.1
0
Urban/Rural
(a)
710,460
A
B
C
D
90,538
282,639
158,651
178,633
12.7
39.8
22.3
25.1
Urban/A urai
(b)
296,004
A
B
C
D
32,624
125,084
67,844
70,452
11.0
42.3
22.9
23.8
4-13
-------
Uridonr Dwttbuton
1 1
10
C
C
7
I
5
4
3
2
0
I I
10
I .
C.
7.
C.
5.
4.
2
1
0
It
10
I
a
7
I
S
4
3
2
t
0
i 23462 76 V1O1112I31415IS171S10S2Iflfl2425
C t C Locton
Figure 3. Indian Cell Landfill Methane Emission Quadrant & 1 . x 1 ..
Grid CIII Locs$ai
Population Cinuo.d Diatritiutlon
4.44
-------
U.1 Qim DIfbu on
40
39
30
20
1 2 3 4 5 8 7 8 9 1O1 121314151S17181P 21 2425
Grid Ciii LocsVon
P ,Iit, n Cirrtro 0 DistrtbuUor
400
350
I ’
1300.
r
10o.
I ________ ________________
0
1 2349S7 S S1O1112t314t51S171S1S2021 392425
Grid C. Lo an
Ut Srn UI*I DlaIrfbu
4O
f 30
II
30.
I .__ __________
1 234517 S S1O11121314151S171S193021 2439
Gnd Ciii LacsIioi
Figure 4. Indian Ca l! Land lIl Methane Emission, Quadrant B. ? x?.
A AP
-------
TABLE 7. INPUT DATA FOR EUROPEAN CELL
Country
Waste Generation Rate
(Mg waste/yr/capita)
Proportion Waste
Landilled’
Austria
0.21
0.70
Belgium
0.31
0.45
France
0.26
0.35
Italy
0.25
0.85
Luxembourg
0.52
0.90
Spain
0.28
0.70
Swltzeñand
0.34
0.90
W. Germany
0.45
0.70
‘Derived from Richards (1989). When specific data not avaliable, proportion was estimated.
-------
and 50 x 5° grids are shown in Table 8; the results for two sets of the 1° x 10 grid are shown in Figures 5
and 6
The results tor the 10° x 10” cell show that, If the Urban/Rural method is closest to the true
emissions, then the Population Centrold method Is the second best. Although it underestimates
emissions In both cells somewhat, the Uniform method overestimates It; furthermore, the magnitude of
the error Is greater for the Uniform method. However, the three methods produce much more simflar
estimates for ‘Europe” cell than tor India 0 cell. This is partly due to the fact that the Europe cell
encloses all or large parts of several densely settled countries. Since these countries are small relative
to the grid, they become, In effect, centrolds. Therefore, less difference between the Uniform and
Population Centrold methods occurs In this cell than In one like India that encompasses pan of a large
country. Finally, the Urban/Rural (b) method shows that iota) estimated emissions are significantly
reduced If a distinction between urban and rural methane generation rates is made. Although these
results show higher emissions for India than for Europe, the methane generation rate used is probably
much too low. A rate four times as high is more likely (Orfich, 1990), making Europe’s emissions more
than double India’s. However, these numbers are for model comparisons only and should not be
treated as true predictions.
When a 5° x 50 grid Is used, considerable disparity can be seen between the three methods,
particularly for India. The Population Centrold method In India produces a very skewed distribution of
emissions in comparison to the Urban/Rural method. For both Europe and India. the Uniform method
comes doses to matching the Urban/Rural method. This Is partly due to the ‘srnaJl country° effect
described above. However, It Is also due to the fact that this particular Population Ceritrold scheme was
designed for a 10° x 10° grId. The formation of centrolds Is based partly on subjective decisions which
are determined in part by scale considerations. Therefore, It is not surprising that centroids developed
at the 10’ x 10’ scale should be too coarse for finer resolutions.
A! the 10 x 10 level, these disparities are even more apparent The Population Centrold method is
dearly Inappropriate at this scale since the relatively few centrolds avalable leave most of the cells
blank. The Uniform method performs somewhat better In that some emissions are allocated to each cell
that is supposed to have emissions. However, the pattern of emissions from the 1° x 10 cells does not
match the Urban/Rural emissions pattern very well.
Based on these results, If It is known that a 10° x 100 grid Is all that Is required, using the existing
Population Centroid methodology Is reasonably accurate and very cost-effective. If maximum flexibility
is desirable or,if a 1° x 1° grid cell Is required, then some new methodology Is required. The
Urban/Rural scheme here could be simplified by using larger metropolitan areas as certtrolds. A similar
approach would be to develop new population centroids, but more of them, and to allocate rural
emissions uniformly.
4-17
-------
TABLE 8. RESULTS OF ALLOCATION ANALYSIS FOR EUROPEAN CELL
Allocation Scheme
Emissions from
1Cxl0 CeII
Emissions tram
5°x5°CeI l
(1&
m 3 CK 4 /yr)
Percent of
10 ’ x 10’
(i0 m 3 CH 4 /yr)
Uniform
438,520
A
B
C
D
59,881
293,601
42,557
42,482
13.7
67.0
9.7
9.7
Centro d
410,475
A
8
C
D
51,394
302,350
20,020
36.710
12.5
73.7
4.9
8.9
Urban/Rural
413,835
A
B
C
D
60,239
258,769
47,484
47,343
14.6
62.5
11.5
11.4
-------
Unitorm D stsibuUon
. 110•
lx.
so
80
70
v.0.
I
I I
1 2 3 4 5 6 7 8 0 1011 12131415161718192021 24
Grid Ciii Loaiion
Po ulalIon C.nlroId. Distribution
110
0 —s—
1 2 3 4 5 I 7 5 9 1011 t21314151I171519 21 V 24
Grid Ciii Location
U b on
110
lix
I S O
! SO
o
5o
[ 40
I 2 flflft _ nr mr 2 lJ
1 2 3 4 5 5 7 $ 5 1011 1213141516i71I19 21 2425
Grid Ciii io siton
Figure 5. European Ceft Landfill Methane Emission. Quadrant B. 1’ x r.
-------
UM im Dl 1bu cn
I
I
F
I ___
1,,
I
I ____
21
1 2 3 4 5 6 7 $ 6 1011 1213141510171$152321 Z2 25
oI c ti iu - .
Figure 6. European Cell Landfill Methane Emission. Quadrant C, f x f.
1 2345678 010111213141518171819202122232425
Grid C&1 Loes
Po uIaton C.nV ida Dii Ibu or
1 23 4517 S •1011121S14151I171$1 52o21 2425
G ca
4-20
-------
4.2.3 Landfills Sampling Program
The greatest gap identified is the lack of sufficient quant i tative data to develop a r&iable model of
methane production. A large number of laboratory studies have been conducted on the microbiological
processes that occur In landfills, but the results are difficult to extrapolate to the field (Senior, 1990). ft Is
also not clear that the theoretical models, which are based on microbial population kinetics models, are
useful for projections at the global scale.
A new approach is needed for the development of a global model. The following pieces of
Information are known:
(1) landfill emissions vary widely;
(2) methane emissions seem to be higher in wetter climates within the temperate zone;
(3) in the temperate zone, air temperature probably does not affect methanogenesis, but in
extreme climates, It may;
(4) the greater the organic content and moisture content of the waste, the greater the methane
potential; and
(5) site and operating characteristics, such as age and depth of landfill, and size of cells, may
affect methane emissions in some way; but, the relative importance or even direction of the
effects are not known.
AEERL Is planning a testIng program that will begin to try to define some of these relationships In
a functional way. Although the scope of the program may eventually become global, In the near future,
a pilot study within the U.S. Is planned. The focus of this study will be to determine whether some
readily available data are sufficient to develop a predIctive model. It Is desirable that elaborate models
requiring very detailed data be avoided due to the expense.
The approach to be used In the pilot study Is to seleCt only landfills where methane Is already
being recovered, Methane content and flow rate will be sampled, analyzed, and the results will be
compared to measurements made by continuous monitors at each site. This wifi allow standardization
of data sets from different sites. Physical characteristics, operating conditions, refuse composition,
refuse acceptance rate, age of landfill, and other pertinent information will be collected by Interviewing
landfill operators.
Climate data from the National Climatic Data Center will be obtained and used with the monItoring
data to determine If any relationship between climate and methane emissions can be found. This
weather data Is collected at airports In major cities; the landfills to be tested will be chosen from those
located near a source of weather dat Sites will also be selected so that a range of climates are
represented; for example, sItes from the southwest, the southeast, and the northeast might be chosen to
represent temperature and precipitation extremes within the contiguous United States.
4-21
-------
5.0 REFERENCES
Bhide. A.. S. Gaiwod, and B. Alone. 1990. Methane from Land Disposal Sites In India. In: Proceedings
of the Irnemationai Workshop on Methane Emissions from Natural Gas Systems, Coal Mining,
end Waste Management Systems. pp 485-502.
Bingemer. H.G., and P.J. Crutzen. 1987. Production of Methane from Solid Waste. J. Geophys. Res.
92:2181-2187.
Bogardus, E. 1987. Landflfls Gas: Asset or Liability? Energy Progress. 7:109-114.
Boolcier, J.T., and R.K. Ham. 1982. Stabilization of Solid Waste in Landfill. J. Environ. Eng. Dry. Am.
Soc. Civ. Eng. 108(EE6):1089.
deWalle, F.B., E.S.K Chian, and E. Hammerberg. 1978. Gas Production from Soild Waste Landfills. J.
Environ. Eng. Div. EE3:415.
Emcon Associates. 1982. Methane Generation and Recovery from Landfills. Ann Arbor Science
Publishers, Inc.. Ann Arbor. Michigan. 135 pp.
Farguhar. G.J.. and F.A. Rovers. 1973. Gas Production During Refuse Composition. Water, Air, Soil
PoUui 2:483.
Jones, HA, and O.B. Nedwell. 1990. Solid Atmosphere Concentration Profiles and Methane Emission
Rates in the Restoration Covers Above Landfill Sites: Equipment and Preliminary Results. Waste
Mgmt. Res. 8:21-32.
Kasali, G.B. 1986. OptimIzation and Control of Methanogenesis in Refuse Fractions. Ph.D. Thesis,
University of Strathclyde, Glasgow, UK.
Klink, E.R., and R.K Ham. 1982. Effects of Moisture Movement on Methane Production In Solid Waste
Landfill Samples. Resour. Recov. Oonserv. 829.
Orlich, J. 1990. Methane Emissions from Landfill Sites and Wastewater Lagoons. In: Proceedings of
the International Workshop on Methane Emissions from Natural Gas Systems, Coal Mining, and
Waste Management Systems. pp. 465-472.
Richards, KM. 1989. LandfIll Gas: Working with Gala. Biodeterioration Abstracts. 3:317-331.
Rodhe, H. 1990. A Comparison of the Contribution of Various Gases to the Greenhouse Effect.
Science 248:1217-1219.
Rovers, FA, and G.J. Farquhar. 1972. Effects of Seasons on Landfill Leachate and Gas Production.
Waterloo Research Institute. Project 8083, Waterloo, Ontario, Canada.
Rovers, FA, and G.J. Farquhar. 1973. Infiltration and Landfill Behavior. J. Environ. Eng. Div. Am. Soc.
Civ. Eng. 99:671.
Rovers, FA, J.J. Tremblay, and G. Moody. 1977. Monitoring and Control. In: Proceedings of
International Seminar. Rep. Eps 4-C-77-4. Fish and Environ. Can. Waste MgmL, Ottawa,
Ontario, Canada.
SCS Engineers. 1986. Gas Emission Rates from SOLId Waste Landfills. Memo to Allen Gesweln,
EPA-OSW. November 17, 1986.
Senior, E. 1990. Microbiology of Landfill Sites. CRC Press Inc.. Boca Raton, Florida. pp. 3-4, 63.
Swartz. A. 1989. Overwiew of International Solid Waste Management Methods. In: State Government
Technical Brief. Paper No. 98-89-Mi -2.
s - i
-------
Versiraste, W., D. Va cke, H. Vaes, and F. Popelier. 1984. Production of Methane by Anaerobic
Digestion of Domestic Refuse and Compost. In: Proceedings of the International Symposium
for Anaerobic Digestion and Carbohydrate Hydrolysis of Waste, Austria.
World Resources InstItute. 1988. World Resources 1988-89. BasIc Books, Inc., New ‘York. pp. 1-372.
t.l
-------
APPENDIX A
WASTE COMPOSITION DATA
The following tables were complied from a variety of sources that are referenced in the footnotes.
Categories of waste have been combined to standardize the data as much as possible. Data are
reponed for 22 countries, but multiple sources were found for several of these. In general, the
industrialized countries are welI.represented with data for 9 European countnes as well as the United
States and Canada. Less data were found for developing nations, particularly South America. the Middle
East, Turkey, Greece, and Indonesia.
A-I
-------
VASTE COMPOSIEIUM IN 0tVtt0Pl* COUNtRIES (WEiGHT PERC(NII
tilt. $ Jakarta , I I
COUNTRY I Indmesla I Indones ia Ind ia Middle East 3 Egypt Egypt
DATA I I I I I
SOURCE 3 A 3 C 3 A I A 3 a 3 C
TEAR I 1904 I I I I
WASTE MATERIAL
Paper/Cardboard 14 * 24 7
Organic Ibusehold Vests 74
V egetab1es PutrescIb1e5 93 08 15
Plast Ics 2 552
Tattles 3.16
Plastics $ leitlies I
Glass I 1 . 7 8 0 1.1 - 2 .0
Metals I 2.08 0 2.3 - 3.8
2.1 - 3 0
Cloth g.s
tcatustibles
Ones 4
Rutbor/Wood/tnther/tlOth 0 2 - I 3
eon. .
Inert below I ui
OtherilWchshhhbtd 4 .34 1 -
A) CMC Press. lOOt Mu. uUUlO9 Of Landf ill Ste ..
8) 0-HateS, W.W., ii at 1188. Mjnlclpsl Solid-Wa ite Management In Egypt. WE IOWA Proceedings of l iii Sifi kitematlonol SO lid Wastes Conference.
C) Qiri, K. tOOL Comprlson of SaNd Wiute Management hi Toinlstlo Macel Developed and Devofoping Counides. W i: IOWA Proceedings of the C lii InternatIonal Solid
Wastes Conference.
20 92- 250
13.0
50 31 1 - 65.6
60 0
25
25
3.0
I
a
9
10
0 131-280 115
D) Pupachiletcu, t tees. Solid Wastes Manegement hi fl iedos . WE IOWA Proceedings of the 5th InternatIonal Solid Wastes Conference.
-------
WAStE COMPOSITION IN DEVELOPING COUNTRIES WE1GHT PERCENT)
CITY, Ist anb ul . Rhodes, 1 10 S I
10 .,anelro
COUNtRY turkey I Greec Ponaco I South Dnerlca Brazil
DATA I I i
SOURCE C I 0 I C ( A
T EAR I 1 I
WASTE MATERIAL
Paper ltar*aard 11.95 11.5 - $5.0 38.40 IS 38.92
Organic Household Waste 32.0 - ISO
VeqetabteslPutresclbles 41.52 35 - IS 55 38 15
Plastics III? 10.0 - 13 0 I 25 6 83
Text Iles 1.28 1.90 3 07
Plastics & tentiles 10
Glass 3.42 10.0 - 22.0 6 41 1 3.19
Hotels 2.2$ 8.5 - 15.0 2 47 8 3.81
Cloth
Cmtusttbles
Fines 14.12 3.3 8
Rubber/Wocd/Leather/Cloth 1.8 - 8.4
Bones
inert below 1 1k m
Other’Unc isnltied 7.49 0.1 - 2.0 1.53 10 8.80
A) CRC Press. 1990. MicrobIology of Landfill Site.
B) El-Natwesi. W.W, at ii. 1908. P*anlcIpst Solid-Waste Management In Fg)pt. in: ISWA Proceedings of the 5th international Solid Wastes Conference.
C) Oaf, K. 1908. ComparIson of $ lid Waste Management In Touristic Pates of Developed and Developing Countries. in: ISWA Proceedings of tI le 5th international Solid
Weetes Conference.
C) Papechr$stou. E. 1008. Solid Wastee Management k Ibodos. Is: ISWA Proceedings of thi 5th InternatIonal Solid Wastes Conference.
-------
VAS1E CONPOSItION IA DEflLOPEO CONtAIU fIIEIGHI P1ACUU)
(Us, I I I I Vendargues %
COUNtRY West GerMany West Germ.np $ Veal 6ennan rrance I Iran e france I
DAtA I I I I 1 - I I
soimc( $ A 8 0 t I
StAR 1911 19* 5 I 1985 I
VASIC NATIRIAI.
Paper $ Cardboard 2? I I 28 2* II 20 - 35
load I lbTd Wastes 2
Organic Household Waste 35 34
Vegetables I Putrescibles
fersentables 15 - 35
te*tile IVood 5 1-6
Plastics. Rubber, Glass. 14 23 12 9 6 6 - 10
leather I teirt le i
Metals S 3 5 6 36 5 -8
lines, Oht Sand. Ash 19 21 31 ID - 20
I Ceramics
8onu
Screenings 10 8 0 .
Other. UnclassIfied sa 4
NoiSturO 7c - 0
A) C Prses. tg a P .Swulitolu5y of landAu Silas.
0) Swirta P1 19*9. OrmM lw 01 kitematiOnd $ I4 Waste M.nagsnisflI Method . State Government Technical BrIef. The ?snedcan Society of Mechanical Englnoa,a,
C) Liw fl . PS. . ml of En. R&D (bio usli) Prograiwni for Landfill Gee. Oep. trnent of Energy, Energy Technology Support Lk fl. Cidoci. England.
0) abed . JO. 19*5. btegfated Meenurco flswvs y , Mu,*lp Waits Processing In Europe: A Status Nopoti on Selected Matedala and ‘orgy Sicovory Prolects. The Wart
Sink, Washington. DC.
E) Cayrol. F.C., et ii. 1988. Ma.roblc Digestion of Municipal Solid Waste by Ihe Valurga Process. in. ISWA Pr Oceedings of die 5th InternatIonal Solid Wades Conlerenca.
-------
WASTE CUNPOSUION S W 8tVtLOP 9 COUNTRIES (WEIGHT PERCENT)
C I I ?. Londen. I $ Wear. I
COUNTRY England England United klrqdcr( (riglend I Sweden Sweden
DAtA I I
SOURCE A 1 R I C 0 I e I nri
1921 5989 198? I 5960 I
WASTE MIUIAI.
Paper $ Cardboard 31.3 33 30 - 40 43 50
Food I Yard Wastes
OrganIc Household Waste
Vegetables $ Putresclbles 23.0 24 iS - 75
Fermentab le .
VeillIes 8 Wood 3 S 4 -
PlastIcs. Rubber. Glass. IS I I ? 12 - IS IS S
leather I leictlles
Neteli 7.7 8 8 9 - I? 6 S
Fines. Dirt. Sand. Mb JO
$ CeramIcs
ScreenIngs I.) 10 iS
Other. UnclassIfied 6 46 10 3 - S 36 U
Moisture 20 - 30 22
A) CT 1C Press. 1990. M obUogy c i Landfill Sites.
B) Sweult, N. 1189. Owervlew 01 international Solid Watts Management Method. Slit. Government Technical Bil.l. The American SocIety of Mechanical Engineer,.
November.
C) La 1 .oon . P.S. i988. The UK Qepsi1me t of Energy RIO (blotuels) Programme toe Land f i ll Gas. Deparimeni *1 Energy. Energy Technology Support Unit. DldcoI , England.
o mert. jo. ig g. i.grete itessuros necovery, ,.,unicipai Processing in Europe A Stalus Rspo,% on Selected Materials end Energy l oovery Projects. The World
Bank, Werhinglon. DC.
E) Cayrol. F.C. •t ci. 1988. Maeroblc Cigeslion of Municipal Solid Waste by the Velorga Process. in- ISWA Proceedings of the 5th Inlernaffonal Solid Wastes Conference.
ri Percent Wet Weight.
-------
V*STt COMPOSItION IN OtVILOP(D COUNtRIES (W(IGNI PERCENT)
CITY. I ICIIYO. I Nadrid
COUNTRY j United Statei United Statai Japan ( Japan I Spain Spain
DAtA I I I I
SOURCE $ A $ S A f 8 8 0
IAR 1978 $ 1981 I ieeo I i a& I i ss I
VASR NATERIAE
Paper I Cardboard 48 37 31.4 IS 19 1
rood I Yard Wastes 23 2 5 32.8
Organic Household Waite 53.6
Vegetables & Putresciblea
Teroontab les
leatfles I Wood 3 3 9
Plastics. Rubber. Glass. II Il 22 12 274
Leather $ Ieattfle,
Metals 9 1.8 2 3 224
unit Dirt. Sand. Ash S 2.3
$ Cerawici
Denes
Scrsenth gs
Other. *McTautfled Ii 50 10 27.3?
Moisture 4$
) CRC Press. 1990 yof Landfill $11.1.
0) Swav . P4. 1989. Overviw of international Solid Waits Management Method. State Government Technical Oust The M erIcan Soclely ol Methanlcst Engineeri.
P1 .
C) Lawson . P.S. egee. TM. UK parbi .nI of Energy R&D Iblofus’s) Programme Me Landfill Gas. Depelment of Energy. Energy Technology Suppoil (Mit. Oldoot. England.
0) Aberl. JO. t985. Wsgrated Resoume Pioewsty . Municipal WiSle Proesisirig OP Slalus PO11 S tected .M*t i 1s md Energy Itocevety Projects. The
Oenh. Washington. DC.
E) Cayrol. F.C.. el ci. 1989. AnaerobIc Digestion of Municipsi Solid Waste by the Valorga Process, in ISWA Proceedings of the 5th internatIonal Solid Wastes Conference.
-------
WAStE COMPOSI V 1ON IN DEVELOPID CSIITflES (wEIGH? PERCE Nt)
CITY. None . I Rate, Vienna,
COIMIRY Ital y Italy - Italy Switzerland I Austria Netherlands
DATA I I I I
SOURCE A I B 0 I I 0 I B
YEAR 1977 1915 I 1980 I 19 85
WASTE MAltREAt
Paper $ Car*oard I ? U 25 30 39 48 23
food 6 Yard Wastes 53
Organic Household Waste 50
Vegetables S Putrescibies 21 82
fenuentab les
texIl es I Wood 2.5 3 8 36
Plastics, Rubber. Glass, 16.5 13 19 22 1705 I i
Leather I Teethes
Metals 3 3 2.5 6 5 )3 3
fines, Dirt. Sand, Ash 7 07
& Ceramics
Bones 109
Screenings
Other. Unclassified S 61 60
Ibisture - - 45 - 50 - .
A) CRC Press. 1989. I saroblology ci tendRil Sitee.
5) Smart, 14. 1989. Overview *1 Mtemational Solid Waste Managamasit EMbed. State Government Technical Brief The Pmerlcan Society of Mechanical Engineers,
November.
C) Lawson. P .S. last The UK Department of Energy R&D (biofuels) P egramme for Landfill Gas Department at Energy, Energy Technology Support Unit, Didool , England.
0) Abe l, JO. 1985. Integrated Risource Recovery. Municipal Waste Proceesing In Europe: A Siatue Report on Selected Materials and Energy Recovery Prolects. me woriri
Bank. Wsshinpton. DC.
E) Ceyrol. cC, et of. last Maerobic Digestion of Municipal Solid Waste by the Valorga Process In ISWA Proceedings of the 5th inlernatlonal Solid Wastes Conlerence
-------
WAStE tONPOSITlOi IN OEVttOPto COUNTRIES IVEIGNI PERCENT)
city. I I I
COUNTRY Canada I israel slang cng
DATA I I I I
SO *Cf I I * $ A I
;;-““i ;;;‘ i - ;;;; “ ;;;““.i..
WASTE NAURIAt
Papr 6 Car oard 36 30.3 5
food & Yard Wastes 51. 5
Organic Nousehold Waste
Vegetables & Putreicibles
Fermentable,
leaShes I Wood 12 2
Plastics. Rubber. Glass. I ? 10
Leather I lealiles
Metals 1 3.1
flnes Dirt. Sand. Ash I?
I Cera.Ica
Sones
Screenings
Other. *ict.ssIfted 45 1.4
Not itere
A cnc Press. 1990. M *rioiogy of Lsndtøh Situ.
B) Swede. P4. tOOB. OvsMsw of NiheffialiOnal Solid Waits MansgIJMld MJ Gd . Stat. Gavmment T.ctrnlcat 84sf The American Society of Mechanical Engineers.
q La, eon . P.S 1995. The UK Department 0$ Energy p.& ,ioiushs) Programme for Land ’ s Gas. Department of Energy, Energy TchnoIogy Support Untt Didcot. England.
0) AbetS. JO. 1995. Wegisted Resource Pecovecy, ,.Iwricip Waste Processing In Europe: A Status Report on Selected Materials end Energy Recovery Projects. The World
PenT ’. Washington, DC.
E) Cayrot. F.C. . 1 si. 1995. Mumble Digestion 0$ Municipal Solid Waits by the Valorga Process. 1n ISWA Proceedings of the 5tl international Solid Wastes Conference.
-------
APPENDIX B
SOUD WASTE MANAGEMENT METHODS IN ELEVEN COUNTRIES
The toHowing data were complied from a report from the American Society of Mechanical
EngIneer Only Industrialized nations were Included in this report. Other sources of data on this
particular aspe of global waste management do edst, however, and more research is probably
warTanted.
‘Swaiiz, & 1989. Ovarview of International Solid Waste Management Methods. In: State Government
Technical Brief. Paper No. 9819-MI -2.
-------
DISPOSAL. 1 1(THOOS OF WASh I II DEVELOPED COWITAIES (WEIGIfT PERCENT)
COUNTRY Vest Gernenp’ France England I Sweden United Stated Japan I
TEAR 1985 1985 119801 1900 1981
DISPOSAL METHODS
lnc lnerat ir,i 28 38 8 53 I D 12
Rec 1ing I D
Lsnd?I llInq $9 41 U II 80 24.6
C ost1ng 8 6 .1
Sorting
Other I I 4 3.3
COUNTRY Spain ltil 7 Switzerland Nettierlandi Canada I
YEAR 1805 1985 1 1980 1685 I 1985
DISPOSAL METHODS
Incineration S 19 80 36 4
Recydflng 2
tandfIlIIng 16 3S 18 55 90
C ostInq 19 5 2 5
SortIng 3 I
Other 30 3 4
Reference: “Overview Of International Solid Waite Managentflt Methods”. State Govermnent Technical Arlef,
Allan Swarti. The hner lcan Soclst of Mechanical Engineers. Noveither. 1989.
-------
APPENDIX C
SAMPLE PREUMINARY L NDF1LL SURVEY FORM
Date: _____________ Person making call: ________________________
Landfill adillty
Name and address: _______________________________________
Contact at landfill: _____________________________________ Telephone:
Methane recovery system in place? ______ Yes ______ No
Active landfill? _____ Yes _____ No Retuse acceptance rate: ________
Date landfill opened: _________________ Closure date: _________________
Describe landfill structure (depth, cell size, etc.): ________________________
If the answer to any at the following questions Is wYes• get a copy at test results It avallablek
Refuse composition known: ____Yes ____No
Results avallable? ____
Moisture tests run on landfill: ____Yes ____No
Results avaIlable? ____
Moisture tests run on refuse? ____Yes ____ No
Results available? _____
Are there perimeter wells or has surface testing of landfill been done? ____Yes ____ No
Results available?
Comments:
C -I
-------
TECHNICAL REPORT DATA
(flcoz read Mwiia:ons on Me reve, c before corn pining!
I REPORT NO
2
3. RECIPIE?4T5 ACCESSIOI4 NO.
EPA- 600 17- 91-002
4. TITLE AND SUBTITLE
5. REPORT DATE
.
Approach for Estimating
Ernis 5 10 1 1 5
Global Landfill Methane
January 1991
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
I. PERFORMING ORGANIZATION REPORT NO.
Rebecca L. Peer. Ann E. Leininger. Barbara B.
Emmel. and Susan K. Lynch
9 PERFORMING ORGANIZATION NAME AND ADDRESS
DCN 90-239-005-48-09
10. PROGRAM ELEMENT NO.
Radian Corporation
P. 0. Box 13000
11. CONTRACT/GRANT NO.
Research Triangle Park,
North Carolina 27709
68-02-4288, Task 48
12. SPONSORING AGENCY NAME AND ADDRESS
EPA. Office of Research and Development
Air and Energy Engineering Research Laboratory
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task final; 4-9/90
14. SPONSORING AGENCY CODE
EPA/600/l3
‘S.Sue;LEMENTARY NOTES AEERL
project officer is Susan A. Thorneloe, Mail Drop 63. 919/
TMACT The report is an overview of available country- specific data and modeling
approaches for estimating global landfill methane. Current estimates of global land-
fill methane indicate that landfills account for between 4 and 15% of the global rnethanl
udget. The report describes an approach for using country-specific and field test
lata to develop a less uncertain estimate of global landfill methane. Development of
nhanced emissions factors for landfills and other major sources of methane will
mprove the understanding of atmospheric chemistry and feedback effects, will target
iltigation opportunities, and will ensure cost-effective mitigation strategies. EPS’ a
rogram to characterize the effects of global change, including identifying and quaxr
fying emission sources, responds to concerns about global warming and is particu
.rly concerned with quantifying emissions sources both in the U. S. and globally.
KEY WORDS AND DOCUMENT ANAI.YSIB
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
C. COSATI FeM/Group
.lution
thane
ission
th Fills
iniating
;te Dispcsal
Pollution Control
Stationary Sources
13B
07C
14G
13C
l5E
TRIBUTIQN STATEMENT
base to Public
I L SECURITY ClASS (Thu Reportj
Unclassified —
21. NO. OF PAGES
53
20. SECURITY CLASS (Thu pmge)
Unclassified
22. PRICE
ivi 2220.1 ($73)
C-2
------- |