EPA-600/R-94-008
January 1994
PROCEEDINGS: THE 1992 GREENHOUSE GAS EMISSIONS AND
MITIGATION RESEARCH SYMPOSIUM
Symposium Chairperson:
Robert P. Hangebrauck (EPA-AEERL)
Acurex Environmental Corporation
4915 Prospectus Drive
P.O. Box 13109
Research Triangle Park, NC 27709
EPA Contract 68-09-0131
Work Assignments 11-25
EPA Project Officer: T. Kelly Janes
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Prepared for:
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
-------
ABSTRACT
The 1992 Greenhouse Gas Emissions and Mitigation Research Symposium, sponsored by the U.S.
Environmental Protection Agency's Air and Energy Engineering Research Laboratory (EPA-AEERL), was
held in Washington, D.C. on August 18-20,1992. The purpose of this symposium was to provide a forum
of exchange of technical information on global change emissions and potential mitigation technologies.
The primary objectives of the meeting were dissemination of technical information and education in recent
research. Oral papers along with an international panel discussion, overheads, slides, and a GloED
demonstration provided for lively exchanges in the following areas: activities in EPA. DOE, and EPRI
on greenhouse gas emissions and mitigation research, and AEERL's global emissions and technology
databases; international activities of selected industrialized and developing countries; carbon dioxide (C02)
emissions and their control, disposal, and reduction through conservation and energy efficiency, and carbon
sequestration including utilization of waste C02; methane (CH^) emissions and mitigation technologies
including such topics as coal mines, the natural gas industry, key agricultural sources, landfills and other
waste management sites, and energy recovery by fuel cells; biomass emission sources and sinks, including
cookstove emissions and control approaches; and energy sources, solar and renewable including renewable
energy options, alternative biomass fuels, advanced energy systems, solar energy developments, and
woodstove emissions and mitigation. The Symposium Proceedings contain 34 submitted papers.
i i i
-------
CONTENTS
Page
SESSION I: OVERVIEW
Frank T. Princiotta, Chairperson
EPA-AEERL
"Greenhouse Warming: The Mitigation Challenge,"
Frank T. Princiotta 1-1
"Methane Reductions are a Cost-effective Approach for Reducing Emissions of
Greenhouse Gases," Kathleen B. Hogan* and Dina W. Kruger 1-25
"Climate Change and Related Activities,"
Kenneth Freidman 1-35
"EPRI's Greenhouse Gas Emissions Assessment and Management Research Program,"
D.F. Spencer* and G.M. Hidy 1-55
"Global Emissions Database (GloED) Software,"
Lee Beck 1-66
SESSION II: INTERNATIONAL ACTIVITIES
Jane Leggett, Chairperson
EPA
"Beyond Rio,"
Hans van Zijst 2-1
SESSION III: C02, EMISSIONS, CONTROL, DISPOSAL AND UTILIZATION •
Ken Freidman, Chairperson
U.S. Department of Energy
"Carbon Dioxide Sequestration,"
Robert P. Hangebrauck*, Robert H. Borgwardt, and Christopher D. Geron 3-1
"The NO A A Carbon Sequestration Program,"
Peter Schauffler 3-14
"The Role of DOE Energy Efficiency and Renewable Energy Programs in
Reducing Greenhouse Gas Emissions,"
Eric Peterson 3-42
"Fuzzy Logic Control of AC Induction Motors to Reduce Energy Consumption,"
R.J. Spiegel*, P. Chappell, J.G. Cleland, and B.K. Bose 3-62
"Methanol Production from Waste Carbon Dioxide,"
Stefan Unnasch 3-71
* Denotes Speaker
V
-------
Page
SESSION IV: EMISSIONS AND MITIGATION OF METHANE AND OZONE PRECURSORS
M.J. Shearer, Chairperson
Global Change Research Center
"Global Atmospheric Methane: Trends of Sources. Sinks and Concentrations,"
M.A.K. Khalil. R.A. Rasmussen, and M.J. Shearer' 4-1
"Coal Mine Methane Emissions and Mitigation,"
David A. Kirchgessner" and Stephen D. Piccot 4-11
"Emissions and Mitigation of Methane from the Natural Gas Industry,"
Robert A. Lott 4-24
"Emissions and Mitigation at Landfills and Other Waste Management Facilities,"
Susan A. Thomeloe 4-46
"Fuel Cell Power Plant Fueled by Landfill Gas,"
RJ. Spiegel* and G.J. Sandelli 4-58
"Methane Emissions from Rice Agriculture,"
M.A.K. Khalil, M.J. Shearer*, and R.A. Rasmussen 4-67
"Livestock Methane: Sources and Management Impacts,"
Donald E. Johnson*. T. Mark Hill, and G.M. Ward 4-81
"Ozone and Global Warming,"
Robert P. Hangebrauck* and John W. Spence 4-85
"Overview of Methane Energy and Environmental Research Programs
in the United Kingdom,"
Suzanne A. Evans. Anton van Santen, Paul S. Maryan, Caroline A. Foster,
Keith M. Richards* 4-97
SESSION V: BIOMASS EMISSION SOURCES AND SINKS
Robert Dixon
EPA
"The Carbon Balance of Forest Systems: Assessing the Effects of Management
Practices on Carbon Pools and Flux,"
Robert K. Dixon*, Jack K. Winjum. and Paul E. Schroeder 5-1
"Global Biome (BlOspheric Mitigation and Adaptation Evaluation) Program,"
Robert K. Dixon* and Jack K. Winjum 5-24
"Agricultural Management, and Soil Carbon Sequestration: An Overview of
Modeling Research,"
Robert B. Jackson IV*, Thomas O. Barnwell Jr., Anthony S. Donigian Jr.,
Avinash S. Patwardhan, Kevin B. Weinnch, Allen L. Rowell 5-32
"Assessment of the Biogenic Carbon Budget of the Former Soviet Union,"
Tatyana P. Kolchugina and Ted S. Vinson* 5-48
"Household Fuels in Developing Countries: Global Warming, Health, and
Energy Implications,"
Kirk R. Smith and Susan A. Thomeloe* 5-61
"The Potential for Energy Crops to Reduce Carbon Dioxide Emissions,"
R.L. Graham 5-81
*Denotes Speaker
-------
Page
SESSION VI; ENERGY SOURCES/SOLAR/RENEWABLE
Robert Williams
Princeton University
"Roles for Biomass Energy in Sustainable Development,"
Robert H. Williams 6-1
"An Analysis of the Hydrocarb Process for Methanol Production from Biomass,"
Yuanji Dong", Meyer Steinberg, and Robert H. Borgwardt 6-26
"Alternative Fuels from Biomass,"
Charles E. Wyman 6-39
"Coproduction of Methanol and Power,"
William Weber*, Arden B. Walters, Samuel S. Tam 6-55
"EPA's Cost-shared Solar Energy Program,"
Ronald J. Spiegel 6-68
"Photovoltaic Developments,"
Jack L. Stone 6-74
"Advanced Energy Systems Fueled from Biomass,"
Carol R. Purvis* and Keith J. Fritsky 6-92
"Programs and Policy Impacts Attributable to Regional Biomass Program
Wood Stove Research Efforts,"
Stephen Morgan 6-99
*Denotes Speaker vii
-------
SESSION I: OVERVIEW
1A
Frank T. Princiotta, Chairperson
GREENHOUSE WARMING: THE MITIGATION CHALLENGE
FRANK T. PRINCIOTTA, DIRECTOR
AIR AND ENERGY ENGINEERING RESEARCH LABORATORY
U. S. ENVIRONMENTAL PROTECTION AGENCY
Human activity has led to an increased atmospheric concentration of certain gases, such
as carbon dioxide, methane, and chlorofluorocarbons, which resist the outward flow of infrared
radiation more effectively than they impede incoming solar radiation. This imbalance yields the
potential for global warming as the atmospheric concentrations of these gases increase. For
example, before the industrial revolution, the concentration of carbon dioxide in the atmosphere
was about 280 ppm and it is now about 355 ppm. Similarly, methane atmospheric
concentrations have increased substantially and they are now more than twice what they were
before the industrial revolution, or about 1.72 ppm. The impact of man's activities is more
dramatic with regard to chlorofluorocarbons. These compounds do not occur naturally; they
were not found in the atmosphere until their initial discernible production several decades ago.
The emissions responsible for increasing concentrations of greenhouse gases are
associated with many human activities, especially the extraction and utilization of fuels, the
large-scale deforestation in many developing countries, and other industrial and agricultural
practices. Our goal at this conference is to discuss the state-of-the-art and the research
opportunities associated with understanding the sources and mitigating releases of these
greenhouse gases. I submit that to understand the mitigation opportunities we need to understand
the fundamental driving forces for releases of these gases.
Let us concentrate for the moment on carbon dioxide, the most important of the
greenhouse gases. The following expression relates the major factors influencing the growth of
carbon dioxide emissions over time for a given country:
FACTORS INFLUENCING GREENHOUSE GAS EMISSIONS
(C02)f = (C02)p x (1 +P+Ip+Ei+Ce)*
where:
(COj), = projected C02 emissions
(C02)p = present C02 emissions
annual growth rate: industrial production
-annual growth rate: population
annual growth rate:
industrial productioi
per capita
annual growth rate:
energy use per unit
unit of industrial
production
annual growth rate: energy use
-annual growth rate, industrial
production
P
annual population growth rate
1-1
-------
Ce
y
The two major factors yielding emission growth are (1) population growth, and (2)
industrialization. These factors can be compensated for by two factors which can mitigate
growth of greenhouse emissions; these are (1) enhanced energy efficiency (i.e., reduced energy
usage per unit of industrial production) and (2) the reduction in the carbon emitted per energy
unit utilized.
It is interesting to note the relative magnitude of these factors expected to influence
emissions of carbon dioxide over the period 1990 to 2025. The Intergovernmental Panel on
Climate Change (IPCC-1992) in their most recent report included a base case of projected
emissions for the greenhouse gases all the way to the year 2100. Table 1 illustrates my
massaging of these data to extract the factors that are important for developed countries
(Organization for Economic Cooperation and Development/OECD countries) versus developing,
or poorer, countries (e.g., Asian countries). The table indicates that, for the developed
countries, the drivers are projected to be primarily economic growth, whereas population growth
is expected to be fairly modest over this time period. The mitigating factors although significant
are projected to be insufficient to counteract population and economic growth, yielding an
estimated 0.7% annual net growth of carbon dioxide emissions from the developed countries.
The situation for the developing countries is even more troublesome. Since their level of
economic activity is currently quite modest, it is projected that they will undergo rapid
industrialization, at the same time that population is growing at a relatively fast pace. Mitigating
factors, namely more efficient use of energy and less carbon intensive energy use, are projected
to be modest over this time period. This yields an expected very large growth of 3.9% annual
increase in carbon dioxide emissions over this period.
Figure 1 illustrates the expected growth in population by area (IPCC, 1992). As you can
see, the developed countries are anticipating relatively low growth rates, whereas the developing
countries, especially Asia, Africa, the Middle East, and (to a lesser extent) Latin America, are
projected to have very large growth rates over the period 1990 to 2100. As stated earlier,
population growth plus rapid industrial growth can yield large increases in carbon dioxide and
other greenhouse gas emissions.
Figure 2 illustrates, based on the expected population and industrial growth, projected
carbon dioxide emissions over the 1990-2100 time frame. The upper graphic in this figure is
the base case for the IPCC 1992 report. It projects growth in emissions from about 7.3 to about
20 gigatons* of carbon over this period, with Asia, Africa, and the Middle East providing much
of the projected growth. The lower graphic is a case I developed to illustrate how important it
is for developing countries to move in a more energy efficient, less carbon intensive path than
have the developed countries during this century. This case was developed by assuming that by
the year 2100 all the developing countries would have a carbon dioxide per capita emission rate
(*) 1 gigaton = 109 metric tons
annual growth rate: carbon emissions i « 1 annual growth rate:
-annual growth rate: energy use J < carbon per unit of
(energy used
years into the future
1-2
-------
TABLE 1.
ASSUMED ANNUAL GROWTH FACTORS INFLUENCING C07
EMISSIONS (1990 - 2025)
(Derived from IPCC, 1992)
FACTOR
OECD
Asia
Growth of Economy Per Capita
2.2%
3.5%
Population Growth Rate
0.3%
1.5%
Growth Rate: Energy Use Per
Economic Output
-1.1%
-0.8%
Growth Rate: Carbon Emissions
Per Energy Use Unit
-0.7%
-0.3%
Annual C02 Growth Rate
(Sum of above factors)
+0.7%
+3.9%
1-3
-------
1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
1OECD minus USA M USA H USSR+E.Europe
mm mm
I Latin America Hi Asia oBAfnca+M.East
Figure 1. Projected population by area (Source: IPCC, 1992).
1-4
-------
BASE CASEJPCC 1992
20
10
40
30
20
Ins lip
II
H-
¦L„ I
1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
¦IOECD minus USA HI USA 1H USSR+E. Europe
i 1 Latin America H Asia S3 Africa+M.East
POOR AREAS EMULATE RICH BY 2100
! 1 ; ! |
• i i !
1
Ml1
ii!,
i : !
i i ;
1 i ¦
' i
! ' !
¦ : ' ; ; 1
i : ! ; i
i ' !
i
i 1 i ¦
; '
Wmm
7 : 1 ¦ . i
1 1 . • i
' I ; i
em IBB «
fflQ |^R9
10
0
1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
mmmm HBBB ¦¦fli
¦ OECD minus USA ¦ USA ¦ USSR+E.Europe
~ Latin America H Asia ^8 Africa+M.East
By 2100 developing countries have
C02 per capita=l/2 current U.S.value=2.9
Figure 2. C02 Emissions: gigatons C for two uncontrolled cases.
1-5
-------
one-half of the current U.S. value, or about 2.9 tons of carbon per capita. Stated another way,
this case assumes that over the next 110 years all developing countries will have a standard of
living and economic structure with an energy use pattern similar to contemporary Europe. In
my view, this is not a wild assumption. Assuming this, emissions are almost twice that
projected by the IPCC. Perhaps this case represents an upper limit of what could happen in the
absence of serious changes in global energy use patterns.
PROJECTED GREENHOUSE WARMING
It is instructive to relate these projected emissions to anticipated global warming. I have
utilized a model I developed which relates emissions of individual greenhouse gases to
equilibrium temperature increases using lifetimes, and radiative forcing functions contained in
IPCC, 1990. Realized (or actual) temperature is estimated using realized projection calculations
in IPCC, 1992, which have been correlated with the model's equilibrium calculations.
Figure 3 relates various emission grown scenarios to projected temperature rise. Note
that this is referred to as realized (or transient) temperature rise, which attempts to take into
account the thermal inertia associated with the Earth's features, especially oceans. (Equilibrium
temperatures, on the other hand, are sometimes reported which ignore the thermal inertia factor.
Typically, these temperatures are 1.5 to 2 times higher than the corresponding realized
temperature increases.) Also note that there are large uncertainties in these numbers, probably
by at least a factor of 2 on both the high and low ends. An atmospheric sensitivity to doubling
carbon dioxide concentrations of 2.5° C was assumed in these calculations. Hie two
uncontrolled emission projection cases we previously discussed were analyzed: the IPCC base
case and what 1 call the fast growth case which assumes that the developing countries will
approach the current industrialized world in terms of carbon dioxide emitted per capita. In the
IPCC base case, warming is estimated at about 3° C by the end of the next century (consistent
with IPCC, 1992). If we were to cap emissions of all greenhouse gases during the year 2000,
it is expected that this warming can be reduced about 30% to a little over 2° C. If the
international community would mitigate further and actually reduce emissions 1 % a year starting
in the year 2000, we can limit the rise to about 1.5° C.
This model calculates that it would require a 2%/yr emission reduction program to
stabilize warming to about 1° C over 1980 levels. I believe that these numbers suggest the major
challenge which faces humankind, if we decide to seriously limit the projected greenhouse
wanning.
A LOOK AT THE IMPORTANT GREENHOUSE GASES
Let us now take a look at the important greenhouse gases and their relative contributions.
Figure 4 shows the projected contributions by the major greenhouse gases assuming the IPCC
(1992) base case over the period 1980 - 2100. As you can see carbon dioxide is the most
important gas with methane and chlorofluorocarbons and their substitutes also important. Note
that the analysis assumed that all countries would reasonably implement international agreements
to phase out chlorofluorocarbons. The model, however, assumes that some of the substitutes,
like HFC-134a, which are substantial greenhouse gases in their own right, will be produced and
1-6
-------
0 1
1980 2000 2020 2040 2060 2080 2100
YEAR
» FAST GROWTH CASE DECLINE EMISSIONS: 1 '/VYR
DECLINE EMISSIONS :2%TVR
Figure 3. Global warming — four cases (control stares: 2000).
IPCC BASE CASE
CAP EMISSIONS 2000
1-7
-------
(60.4%) (C02)
(17.2%)(CH4)
(3.7%) (OTHERS)
(2.8%) (N20)
i6%) CFCs+Sub.
(4.3%) (TROPC. OZONE)
C02
CFCs & Substitutes
METHANE
N20
TROPO.OZONE
! OTHERS
Figure 4. Equilibrium global warning by gas (1st year: 1980; end year: 2100),
1-8
-------
emitted in large quantities.
Note that the contribution of carbon dioxide is probably the best documented of the gases.
The methane calculations, for example, assume indirect effects of methane. Methane
decomposes to other greenhouse gases in the atmosphere: ozone, carbon monoxide, carbon
dioxide, and water. Note that there is a large degree of uncertainty over the magnitude, albeit
not the direction, of these indirect effects of methane. Nevertheless, I included them here since
I believe they ultimately will be important. Also note that other precursors of tropospheric
ozone have many uncertainties as well. There is the need for considerable atmospheric modeling
and measurements to relate emissions of volatile organic compounds, nitrogen oxides, and
carbon monoxide to high-level tropospheric ozone to better understand the significance of this
gas as a greenhouse warming contributor in the upper troposphere.
Figure 5 illustrates the relevant importance of gases over the 100-year time period of the
analysis, again using the IPCC 1992 base case. (Note that the potential cooling effect of
atmospheric aerosols has not been factored into the model calculations.) As you can note, the
short-lived gases, such as methane and ozone, are more important contributors early in this time
frame, with carbon dioxide becoming more dominant later in the time frame. This is associated
with the decay rates of the gases involved. Figure 6 illustrates an important mitigative advantage
in dealing with short-lived gases, such as methane, in that an aggressive control program can
stabilize atmospheric concentrations and mitigate warming relatively quickly. This figure shows
that, if all gases were controlled at 1 % a year from the period 1980 to 2100, essentially all the
methane projected wanning could be mitigated whereas only about 60% of the carbon dioxide
warming could be mitigated since the half-life in the atmosphere of carbon dioxide is so long
that emission reductions don't lead to reduction in atmospheric concentrations until many decades
later. Figure 7 illustrates this phenomenon by plotting concentration ratios relative to 1980 for
two long-lived gases (carbon dioxide and nitrous oxide) and the short-lived methane, all of which
had their emissions reduced by I % a year starting in the year 2000. As you can see, because
of methane's shorter half life, it responds more quickly to mitigation, yielding lower driving
forces for greenhouse warming. Table 2 summarizes what we've discussed relative to the
important greenhouse gases. Note that this table also briefly summarizes major uncertainties
regarding each gas' warming potential, and identifies major human sources.
It is also instructive to estimate the impact of chlorofluorocarbons (and related
compounds). Although these compounds are potent greenhouse gases based on their radiative
properties, recent data suggest that they have been responsible for ozone depletion in the lower
as well as the upper stratosphere. Since ozone in this lower region is a potent greenhouse gas,
there is a net cooling associated with this ozone depletion which opposes the radiative warming
impact. It appears that chlorofluorocarbons have not been the significant greenhouse warming
contributors previously believed. Figure 8 shows the results of model calculations which do not
take into account the cooling effect. What is most interesting is that it is possible that the net
effect of replacing chlorofluorocarbons with compounds such as HFC-134a, which is a potent
greenhouse gas in its own right, could be warming! This could occur since such replacements
which are chlorine free (and therefore non-ozone depleting) will contribute to wanning without
the opposing cooling associated with ozone depletion.
1-9
-------
2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
IMPACT YEAR
CO2 SHH CH4(Direct&Indirect) IB OZONE
CFCs&Substitutes Hi N20 Bi Others
Figure 5. Equilibrium warming , °C, by gas (IPCC, 1992, base case).
1-10
-------
3
C02 CH4 03:xCH4 CFCs N20 OTHERS
H BASE WARMING Zj WARM. PRE VENTED
Figure 6. Warming prevented by 2000 emission control (period: 1980-2100; annual
emission control: 1%/yr).
1-11
-------
1980 2000 2020 2040 2060 2080
__ (C02 PPM)/338 (CH4 PPM)/1.605 N20 PPM/.303
Figure 7. Concentration ratio for three gases (relative to 1980).
1-12
-------
TABLE 2. GREENHOUSE GASES — WHAT IS KNOWN AND WHAT ISN'T
GREENHpUSE GASES _-WH/
iT FS KNOWN A
NgVVHAJ ISN'T
I-. - - .
Characteristic
Carbon Dioxide
Methane
CFC-1 l/CFC-12
IIK -134a
Trojto.()zone
N2(l
1.Atmospheric IJfeHm«(Yra)
50-100
IO:I2.5
65/130
16
<<1
150
{.Current Concentration/
~ 1.26
2.15
Infinite
New CI C Substitute
->1. But Poor Dutu
1.08
Pre-Industrial Concentration
(No Natural Sources)
J.Projected Warming
__ _
22%
_ ... 0.2%
5%
..... 6 Z .
. 3.61
By Gas At the Year 2050
(Incl Indirect EfTects)
(FOR CFC-11 +CFC-I2)
(F.xcludcs CI 14 source)
[Assumes Phaseout)
{.Impact of 1 *Vyr Mitigation:
4Z2_
34%
-
-
6~?i
00 1
"21
Control Starts 2000, the
Impact at 2050: Calculated
as */» of total mitigation:
^-Confidence in Warming
Good
Fair
Poor
Fair/Good
Poor
Fair
Calculations for
Items 3. and 4. Above
-
:
— ¦
i. Major Uncertainties
Quantification of
I.Quantiflc. of Natural,
Recent data suggest
Since no CI, Warming
1 Atmospheric chemistry
Atmospheric
Terrestrial sinks
ftlluman Source s&Sinks
strato.ozone depletion
Impacts not
models insufficient
Concentration
I Explanation needed
may counteract tropo.
Counteracted by
2.Data on Tropo O^one
Rising Faster than
Tor decelerating growth
warming impacts
Strato Ozone Depiction
Trends Poor
known sources/sinks
n Atm Concentrations
}.Emission data for NOx,
predict
Hydrocarbons &CO
—
-
-
- - ¦
Precursors poor
¦
7. Major Human Sources
Fuel Combustion
Coal Mining
Refrigeration Cycles
Refrigeration Cycles
Mobile Sources: VOCs,
Diomass Burning
-Electric Power
Statural $as &. oil
Plastic Foams
NOx and CO
<\dipic&l fN03
-Mobile Sources
prodution and
SolventIJse
Stationary Combustion:
production
-Industrial
transportation
NOx.CO
Mobile Sources
Deforestation
Landfills
Diomass Burning:
harming
Rice Paddies
CO.VOCs
Stationary source
Ruminants
Combustion
Biomass burning
-------
r
U
r»
CO
(L>
feb
(L>
*o
r>
a
::i.
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
-0.1
CFC-11 CFC-12 CFC-113 HFC-22 CC14 134a
PHASEOUT Hi NO PHASEOUT
TOTAL
Figure 8. CFC Equilbrium Warming, °C (1st year; 1980; impact year: 2050; base: IPCC, 1992).
-------
WHAT PENALTIES ARE ASSOCIATED WITH DELAYING MITIGATION?
One important question relative to mitigation is: what penalties are associated with
delaying implementation of a mitigation program? Or stated differently, how much mitigation
opportunity would be lost if a mitigation program were started later rather than sooner? Figure
9 attempts to answer this question by plotting realized (transient) projected warming at 2100 for
three cases as a function of the year the control of all greenhouse gases would be initiated.
Included is a business-as-usual case per the IPCC 1992 emission scenario. Three control cases
are included, two where emissions are reduced by 1 % and 2% a year, respectively; and the third
where emissions are capped at the year control is initiated at that emission level for all
greenhouse gases. Looking at the two decline cases, you can see that a 10-year delay in
initiation of a mitigation program yields a significant diminishment in the ability to mitigate
projected global warming. The graphic suggests that a 20-year delay in a 1 %fyr mitigation
program would require a 2%/yr program to achieve the same degree of warming mitigation that
would have been achieved by the more modest program 20 years earlier.
MITIGATION CHALLENGES FOR THE U.S.
In order to understand the factors influencing greenhouse gas emissions in the U.S., I
have utilized another projection model. This model calculates emissions of carbon dioxide,
methane, and chlorofluorocarbons as a function of input factors such as population growth,
industrial growth, fuel use patterns, energy utilization efficiency, introduction of renewable
energy technology, and mobile source miles per gallon. This model incorporates the electric
utility module described in an earlier paper (Princiotta, 1990).
Figure 10 shows projected equivalent emissions for the 1980-2020 time period for a
business-as-usual case (DOE, 1987). Figure 11 shows the expected increase of carbon dioxide
emissions over the same time period for the major energy/use sectors. This projection suggests
that significant emission increases will result primarily from both increased use of coal to
generate electricity, and growth in the mobile source sector due to a larger auto, truck, and
aircraft fleet. Figure 12 shows that growth in electricity use is a critical parameter in
determining carbon dioxide emissions from the important electric utility sector. Although
introduction of renewable technologies, such as those based on solar or biomass energy sources,
would help mitigate emissions later in this time frame (Princiotta, 1990), the use of so-called
clean coal technologies such as integrated gasification combined cycle (IGCC) will have little
impact. Figure 13 shows that, even with a major introduction of IGCC technology (up to
300,000 MW), only a modest amount of carbon dioxide is mitigated. These results assumed
efficiency for IGCC is 41% vs 37% for conventional coal-fired units; yielding only an 11%
savings in coal use.
Figure 14 illustrates current and projected emissions from the major U.S. sources of
methane. Note the importance of landfills, emissions from cows and sheep, and coal mine and
natural gas pipeline leakage.
Last year the Administration (DOE, 1991) proposed an energy strategy that would have
a significant impact on greenhouse gas emissions. This strategy promoted a major research,
1-15
-------
u
bb
.
a
<
£
2.5
1.5
2000
2010 2020
YEAR CONTROL INITIATES
2030
NO CONTROL(IPCC base case)
MITIGATION: 1%/YR
MITIGATION: EMISS.CAP
MITIGATION: 2%/YR
Figure 9. Warming vs. year control starts, impact year 2100.
-------
V
pi
<
g
rs
O
(J
z
o
H
a
w
i'lll!
- >-
1980 1985 1990 1995
) C0AL/C02
CH4 EQU1V.
2000 2005
0IL/C02 I
CFCs/HCFCs
2010 2015
IGAS/C02
2020
11
10
9
8
7
as
?
6
U4
-------
I
r
00
m
"O
u
i i i H-mi«
IS
\M 'HI' Hit!
1980 1985 1990 1995 2000 2005 2010
"I COAL-ELECTRIC ~ COAL-INDUST.
GAS-RES./COMM
] OIL-OTHER
2015 2020
OIL-TRANSPORT.
GAS-OTHER
Figure 11. U.S. CO2 Emissions by fuel/sector.
-------
(N
C/5
1980
1985
1990
1995
2000
2005
2010
2015
2020
1 % GROWTH 1.5% GROWTH^ 2.0% GROWTH
_b_ 2.3% GROWTH-^ 3.0% GROWTH
Figure 12. C02 Emissions from electricity production (vs. electricity
annual demand growth).
1-19
-------
1980 1985 1990 1995 2000 2005 2010 2015 2020
50000 MW 100000 MW^. 200000 MW^. 300000 MW
Figure 13. CO2 Emissions from electric power (assuming various clean
coal scenarios).
1-20
-------
50
1980 1985 1990 1995 2000 2005 2010 2015 2020
¦ landfills ¦ ruminants ¦pipelines
I • - i COAL MINES ¦ FUEL BURN 1 I OTHER
Figure 14. U.S. Methane emissions (for major sources).
1-21
-------
development, and demonstration program to develop new fuel production and utilization
technologies, with emphasis on renewable and nuclear systems. It also promoted the enhanced
use of natural gas and nuclear power, and the use of renewables as electric utility and
transportation fuels. More efficient utilization of energy, especially electricity, was also
promoted.
The net effect of these policies, if successfully implemented, is shown in Figure 15. The
bottom graphic compares carbon dioxide emissions for the U.S. strategy case with the previously
discussed business-as-usual scenario (upper graphic). Such a policy can approach emissions
stabilized at least over this time interval. As can be seen, the major reduction has been achieved
in the coal-electric sector. This results from a lower growth in electricity demand due to
enhanced end use efficiency, and due to increased use of gas, nuclear, and renewable power
generation displacing carbon-intensive coal-fired power plants.
CONCLUSIONS
Let me summarize what I believe all these graphics and analyses seem to tell those of us
who are interested in greenhouse gas mitigation technology:
(1) Agricultural, medical, and industrial technologies have allowed for unprecedented
population and economic growth; development and application of low-emitting
technology could deal with the potential of unacceptable greenhouse warming.
(2) Technologies and practices could be developed that provide cost effective
mitigation, not just for the developed countries that are generating the bulk of the
emissions in the short term, but also for the developing countries that will likely
be the dominant emitters in the longer term.
(3) Research could help reduce the uncertainties associated with several key gases:
For methane, emission and activity factors need improvement, and the
indirect effects of methane decomposition in the atmosphere needs
clarification. Also the apparent deceleration in the growth of methane
atmospheric concentrations cannot be easily explained by current
source/sink information.
For tropospheric ozone, the mechanisms for formation in the upper
troposphere from the important precursor gases are not fully understood.
For chlorofluorocarbons, the balance between radiative heating and
stratospheric cooling needs to be better understood. Also, results of an
evaluation of likely substitutes for global warming impact would be of
interest.
(4) Carbon dioxide is the key greenhouse gas which is directly linked to fossil fuel
1-22
-------
BUSINESS AS USUAL
fic
<
UJ
fN
W
u
vi
Z
™S
I
1980 1985 1990 1995 2000 2005 2010 2015 2020
9 COAL-ELECTRIC ¦ COAL-INDUST. ¦ OIL-TRANSPORT.
GD OIL-OTHER G~~3 GAS-RES-/COMM.IH GAS-OTHER
C02 PRO/BCTIONS - U.SJiNERGY POLICY CASK
t.
•I
r.
7
1910
1913 1990
I COAL-ELECTRIC
OIL-OTHER
1995
3000 2005
COAL-INDUST
OAS-RES COMM.
2010 2013
I OIL-TRANSPORT.
I GAS-OTHER
2020
9
8
7
6
5
4
3
2
I
0
Figure 15. U.S. CO2 Emissions by fuel/sector.
1-23
-------
combustion, especially coal combustion. Major efforts could provide an
alternative energy path emphasizing renewable technologies such as solar, and
biomass with a focus on electric power production.
(5) Methane provides a ripe opportunity for mitigation research. By controlling the
major human sources, such as coal mines, landfills, and natural gas pipelines,
methane atmospheric levels could be stabilized within a relatively short period,
with substantial near-term mitigation impacts.
REFERENCES
Intergovernmental Panel on Climate Change (IPCC), "Climate Change 1992 - The
Supplementary Report to the IPCC Scientific Assessment," 1992
IPCC "Scientific Assessment of Climate Change," June 1990
Princiotta, F.T., "Pollution Control for Utility Power Generation, 1990 to 2020," Proceedings
of the Conference: Energy and the Environment in the 21st Century, March 26-28, 1990, MIT
Press, Cambridge, MA
U. S. Department of Energy, "Energy Security, A Report to the President of the U.S.," April
1987
U. S. Department of Energy, "National Energy Strategy," February 1991
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency s peer and
administrative review policies and approved for presentation and publication.
\
1-24
-------
METHANE REDDCTI0N8 ARE A COST-EFFECTIVE APPROACH
FOR REDUCING EMISSIONS OF GREENHOUSE GASES
Kathleen B. Hogan
Dina W. Kruger
Office of Air and Radiation
U.S. Environmental Protection Agency
Washington, D.C.
1-25
-------
Methane reductions can play a large role in providing low
cost, if not profitable, opportunities for reducing
greenhouse gas emissions, in addition to other benefits for
the atmosphere. This paper reports on opportunities for
reducing emissions of methane which have been examined
through a number of EPA efforts and activities developed by
the Intergovernmental Panel on Climate Change.
BACKGROUND: METHANE IS AN IMPORTANT GREENHOUSE GAS
• Methane concentrations are currently rising in the earth's
atmosphere and continuing increases have been projected for the
future [1,2,3]. These increases are largely correlated with
increasing populations, and currently about 60 percent of global
methane emissions are associated with human activities (Fig. 1).
The benefits of reducing methane emissions from these
anthropogenic sources will be substantial for several reasons.
First, methane is a potent greenhouse gas and reductions in
methane emissions would be 2 0 to 60 times more effective in
reducing the potential warming of the earth's atmosphere over the
next century than reductions in CO, emissions [2]. Second,
methane released from human activities is a wasted resource, and
these activities can likely be redesigned to profitably benefit
from the efforts taken to reduce the methane emissions. Third,
reductions x&et^iane gidjlssxohs wxlX providifi l^enefi^*s of red\icn^g
the risks of increasing tropospheric ozone and reducing the
earth's oxidizing potential [3,4,5,6,7]. Stabilizing CH6
concentrations may reduce expected global tropospheric 0,
increases, and while more uncertain, may allow OH to return to
about current levels by 2100 after a significant suppression [8].
Importantly, relatively small reductions in methane emissions of
40 to 60 Tg/yr, or about 10 percent of annual emissions, can halt
the annual rise in methane concentrations. This assumes that the
rate of CH4 destruction (by OH) remains the same [2,3,8].
Due to the potency of methane in the atmosphere and its
relatively short lifetime, stabilizing methane concentrations may
act to substantially reduce potential warming. The results of
holding methane emissions at about constant levels (i.e., 500 Tg)
throughout the next century are shown in Fig. 2 (examined using
the IPCC scenarios for future emissions of the greenhouse gases
and the Atmospheric Stabilization Framework (ASF), a model used
1-26
-------
in support of the IPCC [ 3 ]) . The expected warming over the next
century is reduced by about 1°C or 25 percent of the expected
warming post 1990. This reduced warming is similar to the
reduced warming which would result from stabilizing CO, emissions
at 1990 levels (while concentrations continue to rise to over 500
ppm), and the reductions in warming from stabilizing CH4
concentrations or C02 emissions are virtually identical through
the year 2050 [9].
OPTIONS FOR REDUCING METHANE
Opportunities for reducing methane emissions from its major
anthropogenic sources have been identified and reviewed through
EPA efforts and expert meetings held under the IPCC [4,10,11].
In total, it currently appears to be technically feasible to
reduce methane emissions on the order of 120 Tg/yr (75 to
170 Tg/yr). Over the next 10 years about one-third to one-half
of these reductions would be needed in order achieve the
necessary 40 to 60 Tg reduction to stabilize atmospheric methane
concentrations.
The options for reducing methane emissions include:
• Ruminants: Ruminants world-wide are likely the second
largest anthropogenic methane source, emitting 65 to 100 Tg
[12]. Methane emitted from ruminants is a lost opportunity
to transform more carbon into useful product such as meat or
milk during the natural fermentation of feed. In developed
countries, specific feeds have been identified which may
reduce methane emissions while enhancing the productivity of
the cattle. Additionally, in the United States,
administration of bovine somatotropin (bST), provided it
receives regulatory approval, could increase milk production
while reducing methane emissions on the order of 10 percent
[personal communication]. In other regions of the world,
programs to increase animal productivity through strategic
supplementation (i.e., the use of supplements such as
molasses urea blocks to address livestock nutrient
deficiencies) have been initiated. These programs have
cost-effectively increased milk production and reproduction
efficiency while reducing methane emissions by perhaps as
much as 60 percent per gallon of milk [13].
• Animal wastes and wastewater treatment: Animal wastes may
contribute about 20 to 30 Tg [14] of methane and wastewater
treatment an additional 25 Tg [8]. Methane recovery systems
can profitably capture 50 to 80 percent of the methane
emitted by anaerobic waste management lagoons, as
demonstrated by systems around the world. Such lagoons are
used primarily at dairy and swine operations and account for
over one-third of total emissions from animal wastes.
1-27
-------
Recovered methane may be used directly or used to generate
electricity [15]. Opportunities to employ methane recovery
systems exist in the United States and many other countries.
Rice cultivation: Rice cultivation may be the largest
anthropogenic methane source. Over the long term, methane
emissions can likely be reduced by 10 to 30 percent by an
integrated management approach to irrigation, fertilizer
application, and cultivar selection. Additional research is
necessary to develop management practices for rice growing
that will maintain rice productivity and reduce methane
emissions [16].
Biomass burning: Methane emissions resulting from biomass
burning to clear new lands and after cropping and for energy
purposes are estimated to range from 20 to 80 Tg. These
emissions can be reduced through fire management programs,
encouraging the use of alternative agricultural practices,
and the use of more efficient cookstoves [12,17,18].
Landfills: Landfills currently account for about 20 to 70
Tg of methane from the anaerobic decay of wastes [12],
primarily from developed countries. Recovery systems can
reduce emissions by 50 to 90 percent in existing landfills
by collecting the medium BTU gas. The recovered gas may be
burned directly in nearby industrial boilers or used to
generate electricity, or at a minimum flared. The first two
options could serve to displace C02 emissions from fossil-
based fuels. Commercial operations in the United States,
Western Europe, and other regions show that these systems
are operated profitably. Furthermore, the United States is
proposing rules to reduce emissions of air contaminants
(primarily nonmethane organic compounds and air toxics) from
landfills. As a side benefit methane emissions from U.S.
landfills will be reduced by about 40 to 80 percent
depending upon the stringency of the final rule [19].
coal mining; Coal mining accounts for 30 to 50 Tg of
methane with most of the methane emitted from a small number
of highly gassy mines [20]. Degasification using vertical
wells in advance of and during the mining operation can
reduce emissions of methane trapped in gassy underground
mines by more than 50 percent while reducing the costs of
necessary mine ventilation [8,11]. A portion of the
recovered gas can be high BTU gas and fed directly into a
pipeline or used to generate electricity. Gas of medium BTU
quality recovered from gob wells over the mining operation
(in the gob area) may be used to generate electricity. Many
opportunities exist to cost-effectively expand methane
recovery and use at coal mines in the United States, Poland,
Czechoslovakia, the former Soviet Union, and the People's
Republic of China, as well as elsewhere [21,22,23]. In the
1-28
-------
future, emissions may be reduced further by using mine
ventilation air that contains less than one percent methane
as combustion air in gas fired turbines as is currently
being done in Germany and being considered in Nova Scotia.
Oil and natural gas systems! Oil and natural gas systems
account for about 30 to 70 Tg of methane [12]. Most of the
emissions are likely outside of the developed countries,
with significant contributions from countries with economies
in transition such as Eastern Europe and Russia. Improved
handling of casing gas during oil production will reduce
venting and flaring of natural gas during oil production.
Improved technologies such as low bleed pneumatic
instruments, high integrity piping, and new pipeline
restoration techniques can also cost-effectively reduce
methane emissions [8]. In addition, emissions from gas
transmission and distribution in the USSR (where leakage has
been estimated on the order of 5 percent of throughput) and
Eastern Europe could be reduced substantially by improving
these facilities [personal communication].
EFFORTS REQUIRED TO ACHIEVE THESE REDUCTIONS
These options for reducing methane emissions offer
substantial reduction potential, but have not been implemented on
a wide scale to date. A number of barriers currently limit the
implementation of these options around the world. This includes
financial, informational, political, and in some cases technical,
barriers.
Future efforts should focus on removing these barriers. For
example, in the United States, it can be profitable to recover
and utilize methane that would have been emitted from coal mines
due to the value of the methane and reduced ventilation costs,
but institutional questions of methane ownership and constraints
on receiving fair prices for gas or electricity may block
implementation. These legal and pricing issues need to be
resolved. Similarly, strategic supplementation of livestock can
substantially increase livestock productivity and create a local
market for supplementation crops, but lack of capital and
infrastructure may block implementation in many countries.
In many cases, technology demonstrations will be crucial to
removing existing barriers. Upon the clear demonstration that
certain technologies have substantial pay backs and large
environmental benefits, much more energy will be placed toward
removing other barriers.
y-29
-------
SUMMARY AND CONCLUSION
Controlling methane emissions appears to be technically
feasible and cost-effective with current technology. These
reductions in CH, emissions appear to be very effective in
mitigating radiative forcing in the atmosphere and eventual
global warming. Reduction in CH6 emissions will also likely
moderate large-scale increases in tropospheric 03 and to counter
the suppression of OH.
Efforts to encourage methane reductions will require
identification and removal of a number of barriers which hinder
the implementation of available technologies. These efforts will
need to include well-planned demonstration projects, in many
cases.
REFERENCES
1. Blake, D.R. and F.S. Rowland, " Science. 239: 1129-1131,
1988.
2. Shine, K.P., R.G. Derwent, D.J. Weubbles, and J-J Morcrette,
"Radiative Forcing of Climate," Climate Change: The IPCC
Scientific Assessment. Intergovernmental Panel on Climate
Change, J.T. Houghton, G.J. Jenkins, and J.J. Ephraums eds.,
Cambridge University Press, Cambridge, 1990.
3 • U.S. EPA, fPr S^t^tiliiliZilin^f CilifflBt'? /
Draft Report to Congress, D. Tirpak and D. Lashof eds., U.S.
Environmental Protection Agency, Washington, D.C., June,
1990.
4. crutzen, P.J. The Geophvsiolocrv of Amazonia. R.E. Dickinson,
ed., John Wiley, New York, pp. 107-130, 1987.
5. Thompson, A. M., R. W. Stewart, M. A. Owens, and J. A.
Herwehe, Atmospheric Environment. 23: 519-532, 1989.
6. Watson, R.T., H. Rodhe, H. Oeschger, and U. Siegenthaler,
"Greenhouse Gases and Aerosols," Climate Change: The IPCC
Scientific Assessment, Intergovernmental Panel on Climate
Change, J.t. Houghton, G.J. Jenkins, and J.J. Ephraums eds.,
Cambridge University Press, Cambridge, 1990.
7. Dak Sze, N. " Science. 195: 673-675, 1977.
8. intergovernmental Panel on Climate Change, Methane Emissions
and Opportunities for Control. Workshop Results of the
Response Strategies Working Group, September 1990.
1-30
-------
9.
10
11
12
13
14
15
16
17
18
19,
20.
Hogan, K.B., J.S. Hoffman, and A.M. Thompson,3"Methane on
the Greenhouse Agenda," Nature. Vol. 354, November 21, 1991.
Subgroup on Agriculture, Forestry, and Other Human
Activities, Greenhouse Gas Emissions from Agricultural
Systems. Response Strategies Work Group, Intergovernmental
Panel on Climate change, U.S. EPA, September, 1990.
Energy and Industry Subgroup, Energy and Industry Subgroup
Report. Response Strategies Work Group, Intergovernmental
Panel on Climate Change, U.S. Environmental Protection
Agency, May, 1990.
Intergovernmental Panel on Climate Change, Climate Change
1992; The IPCC Supplement Report. 1992.
Leng, R.A. Improving Ruminant Production and Reducing
iEiiJiSPipn? frpin ^y.
Supplementation. U.S. Environmental Protection Agency,
Washington D.C., June, 1991.
Safley, Jr., L.M., M.E. Casada, J.W. Woodbury, and K.Roos.
Hsthsns Emissions from 9nd PQMXtry Msnvirs»
U.S. Environmental Protection Agency, Washington, D.C.,
1992.
K. Roos, Profitable Alternatives for Regulatory Impacts on
WsnfliSffiSlSnIr» Proceedings of American
Society of Agricultural Engineers 1991 National Livestock,
Poultry, and Aquaculture Waste Management Meeting.
U.S. EPA, Sustainable Rice Productivity and Methane
Reduction; Research Plan. K. Hogan and B. Braatz eds., U.S.
Environmental Protection Agency, Washington, D.C., May,
1991.
Crutzen, P.J., I. Aselmann, and W. Seiler. Tellus. 38B: 271-
284, 1986.
U.S. Agency for International Development, Greenhouse Gas
Emissions and the Developing Countries; Strategic Options
and the U.S.A.I.D Response. Report to Congress, July 1990.
Federal Register, United States Government Printing Office,
Vol. 56, No. 104, pp. 24468-24528, 1991.
Boyer, C. M., J.R. Kelafant, V.A. Kuuskraa, K.C. Manger, and
D. Kruger. Methane Emissions from Coal Mining; Issues and
Opportunities for Control. U.S. Environmental Protection
Agency, September 1990.
1-31
-------
21. Pilcher, R.C., et al., Assessment of the Potential for
Economic Development and Utilization of Coalbed Methane In
Poland. U.S. Environmental Protection Agency, August 1991.
22. Bibler, C.f et al.f Assessment of the Potential for Economic
Development and Utilization of Coalbed Methane in
Czechoslovakia. U.S. Environmental Protection Agency, (in
press).
23. Huang, J.P., Opportunities for Coalbed Methane Recovery and
Utilization in China: The Potential for U.S.-China
Cooperation, U.S. Environmental Protection Agency, September
1990.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency s
peer and administrative review policies and approved for presentation and publication.
1-32
-------
Methane Emissions Sources
Human
Activities
Account for
-60% of
Emissions
Livestock
Animal Waste
Rice
I
Landfills
Coal Mining,
Natural Gas and
Petroleum Industry
Blomass Burning
Natural Systems
20 30
Total Emission
400-600 Tg
300
t—i—i—i—i—I—i—i—i—r
20 40 60 80 100 Emissions Estimate (Tg)
Slnca IPCC (based on NOAA work)
range of estimates J— H additional uncertainty
B08023 7
-------
5.0.
Actual
Temperature
Increase ('C) 3_0 _
Roughly Identical effects on actual warming
- C02 emissions stabilization
- CH4 Concentration stabilization
o
1988
t
t
2000
2025 2050
Year
2075
Assumes 3° equilibrium warming
] Constitutes uncertainty range due to NOx
IPCC-BAU
CH4 stabilization
C02 capped at 1990
CH4 and C02
2100
Figure 2.
Benefits of methane stabilization where methane emissions are capped at 540 Tg/yr as compared
to capping C02 emissions at 1990 levels (and concentrations grow to over 500 ppm by 2100)
-------
Climate Change and
Related Activities
Kenneth Freidman
Department of Energy
Washington, D.C.
U.S. Department of Energy
¦ mm *
1*35
-------
Among all the challenges in our tenancy of this planet, climate change is.
of course, foremost in our minds. We're leading the search for response
strategies and working through the uncertainty of both the science and the
economics of climate change. But there is one area where we will allow
for no uncertainty—and this is our commitment to action—to sound
analysis and sound policies.
President Bush. White House Conference
on Science and Economics Research
Related to Global Change
1-36
-------
Foreword
The production and consumption of energy contributes to the
concentration of greenhouse gases in the atmosphere and is the focus of
other environmental concerns as well. Yet the use of energy contributes to
worldwide economic growth and development. If we are to achieve
environmentally sound economic growth, we must develop and deploy
energy technologies that contribute to global stewardship.
The Department carries out an aggressive scientific research program to
address some of the key uncertainties associated with the climate change
issue. Of course, research simply to study the science of global climate
change is not enough. At the heart of any regime of cost-effecuve actions
to address the possibility of global climate change will be a panoply of
new technologies—technologies both to provide the services we demand
and to use energy more efficiently than in the past. These, too, are
important areas of responsibility for the Department.
This report is a brief description of the Department's activities in
scientific research, technology development, policy studies, and
international cooperation that are directly related to or have some bearing
on the issue of global climate change.
James D. Watkins, Secretary of Energy
1-37
-------
Table of Contents
Introduction 1
The Earth's Climate 1
DOE's Role: Historical Background and Issues 1
Part I: Focused Activities 3
Scientific Research 3
Computer Modeling 3
Earth Systems Research 3
Carbon Cycle Research 3
Atmospheric Research: Atmospheric Radiation Measurement 3
Oceanic Research 4
Terrestrial Research 5
Outreach. Communication, and Exchange of Data 5
Policy Analysis 5
International Activities 6
Pan II: Related Activities 7
Scientific Research 7
Transportation 7
Alternative Fuels Development 7
Vehicle Development 7
Residential. Commercial, and Industrial Efficiency 8
Building. Appliance, and Equipment Efficiency 8
Industrial Process Technologies 9
Industrial Waste Minimization 10
Electricity Generation and Use 10
Integrated Resource Planning 10
Electricity Generation II
Policy Analysis 13
National Energy Strategy 13
Alternative Fuels 13
Conservation and Efficiency Studies 13
Additional Studies 13
Related International Activities 15
International Agreements 15
Bilateral and Multilateral Cooperation 15
Promoting U.S. Exports 15
Working Group of the DOE Climate Change Executive Committee 17
OOE Climate Chang® and Related Activities
1-38
-------
Introduction
The Earth's Climate
The Earth's atmosphere contains gases that absorb, and
then radiate back to Earth, some of.the Earth's thermal
energy that would otherwise be radiated to outer space.
These gases result from both natural and manmade
processes and include carbon dioxide, nitrogen oxides,
methane, chlorofluorocarbons, halogenated compounds,
water vapor, and others. They are referred to as
"greenhouse" gases because their action is somewhat
like the process that makes it possible for a greenhouse
to capture the heat of the Sun and maintain warmer
temperatures during winter weather.
This "greenhouse" process regulates the Earth's climate
at a level to sustain life, making our planet unique. The
term "climate" refers not only to temperature, but also
to the entire system of precipitation, cloudiness, and
winds, as well as to the distribution of these features in
space and time.
Human activities, in addition to natural processes, affect
concentrations of greenhouse gases. These activities
include energy production and use, agriculture,
industry, and deforestation. It is the effect of these
increasing manmade emissions that is the topic of
current research and debate. Simply put. the research
hypothesis is that manmade greenhouse gases in the
atmosphere could cause changes in the Earth's climate.
Figure 1 shows the contribution of different gases to the
potential for warming (called radiative forcing).
The scientific community is unsure about how the Earth
responds to increases in manmade greenhouse gases.
Does the ocean have processes that can absorb these
extra greenhouse gases? Do clouds act as a natural
buffer to regulate temperature changes? What kind of
climate changes could occur, where, and when? These
are the questions that scientists around the world are
attempting to answer.
DOE's Role: Historical Background
and Issues
Global climate change is a significant issue for the U.S.
Department of Energy (DOE) because greenhouse gases
are emitted from the production and use of fossil fuels.
Energy use and production now contribute more than
half of the total manmade emissions on a global basis.
Figure 2 shows an estimate of the relative contributions
of various sources to total manmade emissions of
greenhouse gases.
Our choice of energy sources can affect the emissions
of greenhouse gases because some energy sources emu
more greenhouse gases than others. Conversely,
changes in climate could affect energy systems and
energy demand. DOE is pursuing a wide range of new
technologies that can help reduce future greenhouse gas
emissions. These efforts include higher efficiency,
cleaner electricity production, and a variety of
Figure 1. Greenhouse Gas Contributions
to Radiative Forcing
70%
40% -
30%-
20%-
10% -
CFC's
Source: IPCC Scientific Assessment
Note: The contribution from tropospheric ozone may also be
significant, bui cannot be quantified at present.
DOE Climate Change and Related Activities
1-39
-------
renewable energy options. In addition, the Department
is developing technologies for energy efficiency in
buildings, industry, and transportation.
In 1977. the National Academy of Sciences challenged
(he scientific community on the subject of energy and
climate by stating: "To reduce uncertainties and .to
assess the seriousness of the matter, a well-coordinated
program of research that is profoundly interdisciplinary
in character, and strongly international in scope, will be
required."
Responding to this challenge and the growing concern
about the long-range consequences of carbon dioxide
emissions resulting from ever increasing fossil fuel
combustion prompted the Department of Energy to
undertake a thorough examination of the effects of
carbon dioxide (CCM emissions. The first step was to
convene a Workshop on the Global Effects of Carbon
Dioxide from Fossil Fuels (1977, Miami Beach. FL).
Some 75 scientists discussed the state of knowledge of
the C02 cycle and the consequences of increases in
atmospheric C02. The workshop identified significant
gaps in understanding and recommended actions to fill
these gaps. This led to the development of the Carbon
Dioxide Research Program at the Department of Energy
in 1978.
The 14 years of research since then by the Carbon
Dioxide Research Program has laid the basis for the
recent science assessment by the Intergovernmental
Panel on Climate Change. Accomplishments along the
way have included the global C02 emissions inventory,
the global temperature data base used to assess climate
change over the last 130 years, leading the diagnosis
and improvement of climate models, establishing the
C02 vegetative fertilization effect as a major beneficial
impact, leading the development of a ground-based,
remote sensing network to determine the role of clouds
in climate change, and initiating the research to
incorporate emerging supercomputer hardware and
software into climate model development. The Program
continues to provide scientific leadership on climate
modeling, atmospheric, terrestrial, and oceanic data
collection, measurement, and analysis.
The Department also established a Global Climate
Change Executive Committee, jointly chaired by the
Deputy Under Secretary for Policy, Planning and
Analysis
-------
Part I: Focused Activities
Scientific Research
The Carbon Dioxide Research Program conducts the
Department's focused scientific research on climate
change. The goals of the program are to estimate the
future levels and rate-of-increase in atmospheric carbon
dioxide (CCM and other energy-related emissions and
to understand and predict potential effects of emissions
on climate and biota. This information is required to
scientifically underpin energy policy options aimed at
preventing, mitigating, or adapting to increasing
greenhouse gas concentrations and global environ-
mental change. Major activities include < 1) developing
computer models to predict rate and magnitude of
global ana regional climate change; (2) understanding
the systems that control the current and past climates of
the Earth: and (3) exchanging and communicating data
and modeling results with other climate researchers
around the world.
Computer Modeling
Complex computers are used as tools to predict future
climate trends. The Department has extensive climate
modeling capabilities using advanced supercomputers.
However, additional computer power and an advanced
climate model (ACM) will be needed to further
improve climate modeling and prediction efforts.
The Computer Hardware. Advanced Mathematics and
Model Physics (CHAMMP) research program is a 1991
initiative of the Department of Energy's climate
modeling program. The objective of the program is to
accelerate and improve the global and regional
predictive capability of climate models. This requires
an ACM capable of more detailed simulations over
longer time intervals and incorporates significant
improvements in the representation of the physics and
chemistry of the climate system.
CHAMMP addresses these challenges in a phased
approach over ten years. In the near term (2 to 3 years),
its goals are to improve performance of existing climate
system models by taking advantage of emerging
parallel computers. In the intermediate term (3 to 6
years), the CHAMMP program aims to develop initial
versions of ACM systems that are capable of better than
100 gigaflop (billion floating point operations per
second) performance. In the long term (6 to 10 years),
DOE Climate Change and Related Activities
CHAMMP wjll improve the initial versions of the
ACM components, assemble an optimized ACM. and
begin detailed research caicuiations on the faster
supercomputers.
Earth Systems Research
Carbon Cycle Research
Carbon cycle studies seek to better understand the
different sources and sinks of carbon. A "source" emits
carbon (in the form of C02) into the atmosphere: a
"sink'" absorbs and stores carbon. For instance, trees are
primarily made up of carbon absorbed from the
atmosphere. Therefore, forests are considered sinks of
carbon during their prospective lifetimes. As they decay
or are burned, carbon is released back into the atmo-
sphere in the form of COi and the tree becomes a source.
There are other sinks of carbon, including surface soils
and the surface of the ocean, about which little is
known. The Department conducts and supports carbon
cycle research because a better understanding of natural
sinks and sources of carbon will help to determine the
extent of increased atmospheric concentrations of
manmade greenhouse gases. Carbon cycle research is a
common thread in the fabric of DOE's atmospheric,
oceanic, and terrestrial research.
Atmospheric Research: Atmospheric
Radiation Measurement
The Atmospheric Radiation Measurement (ARM)
program has three main goals. The first is to measure
and describe the radiation balance (natural greenhouse
effect) from the Earth's surface to the very "top" of the
atmosphere. The second is to understand the role clouds
play in this process, so that climate models can be
improved. The third is to assist in verifying satellite
measurements with ground-based data.
The program will accomplish these goals by
establishing five observing stations around the world.
These sites will be selected to provide a worldwide
view of differing cloud and atmospheric conditions. The
sites will employ specialized ground-based remote
sensing instrumentation, specialized aircraft, and
balloon platforms to collect and analyze data. The first
ARM site will be established in 1992.
3
1-41
-------
Oceanic Research
The ocean research objectives are to produce a global
survey of carbon dioxide in the ocean, to better
understand the ocean's role as a source and sink of
carbon dioxide, and to improve ocean circulation
models for climate research. To meet these objectives.
DOE supports laboratory, observational, and modeling
studies to understand the mixing, transport processes,
and carbon cycling in the ocean, and the exchange of
heal and carbon between the ocean and the atmosphere.
A better understanding of these issues is necessary to
estimate the ocean's uptake of carbon dioxide produced
by fossil fuels.
At-Sea Data Collection and Analysis
The Department cooperates with other Federal and
international agencies in collecting ocean data. The
World Ocean Circulation Experiment is coordinating a
global survey of oceanic carbon dioxide. Specialized
research ships collect data from both the Atlantic and
Pacific Oceans. DOE continues to develop the
instrumentation needed to collect and analyze oceanic
carbon dioxide data.
Remote Sensing: "Smart Buoys" and the
Heard Island Experiment
At-sea data collection could be complemented by a
device now under development by the Department.
Expendable Ocean Sensors or "smart buoys" are remote
sensing devices that can submerge to the ocean floor to
collect ocean temperature, salinity, and density data on
their descent. At a programmed time interval, the buoys
ascend to the ocean's surface, collecting the same data
during ascent. Once at the surface, the buoys transmit
the data to a satellite. These devices could verify
satellite observations and provide new data for the next
generation of ocean models. Figure 3 shows a diagram
of such a buoy.
The objective of the Heard Island Experiment is to
assess the feasibility of using sound as an ocean
thermometer. The speed of sound in seawater is
determined by water temperature. The travel time for a
sound signal to cross the ocean is therefore proportional
to the average temperature of the ocean along the sound
path. The Department's Environmental Sciences
Division, in cooperation with the Office of Naval
Research, the National Science Foundation, the
National Oceanic and Atmospheric Administration, and
10 foreign countries have supported the Scripps
Institute of Oceanography in executing the Heard Island
4
Figure 3. Diagram of a "Smart Buoy"
Remote Sensing Device
DOE Ciimate Change and Related Activities
1-42
-------
Experiment. The experiment transmitted a low
frequency sound from a ship near Heard Island in the
south Indian Ocean. Sensors around the world were
used to detect the signal. An assessment of the
experimental data is expected in late 1992.
Terrestrial Research: Vegetation
The Department's unique research on plant response to
variable CO-, concentrations will focus on plant carbon
metabolism in response to higher concentrations of
CO-,. Field experiments have shown dramatic biological
responses to elevated levels of carbon dioxide. The free
Air COt Enrichment (FACE) experiment is an open-air
control system in which plants are exposed to high
concentrations of C0->. Cotton was first used to test the
FACE system, and the experimental results were
striking. Plants grown at elevated CO-> produced a
marketable cotton yield 5 weeks earlier than usual, and
wuh 50 percent more biomass.
These results have profound implications for the
capacity of vegetation to fix and sequester atmospheric
CO-), and thus possibly slow the rate of atmospheric
COn increase and its possible effect on global climate.
Outreach, Communication,
and Exchange of Data
Carbon Dioxide Information Analysis Center
The Department provides climate data and information
services to the domestic and the international scientific
communities. The Carbon Dioxide Information
Analysis Center (CDIAC) at Oak Ridge National
Laboratory is a public repository of greenhouse gas data
and information. CDIAC maintains format and quality
control standards and catalogs diverse data sets to
provide a uniform data library and easy conversion to
modeling systems so that researchers can easily access
and characterize the data.
For more information please contact:
Oak Ridge National Laboratory
Carbon Dioxide Information Analysis Center
Building 1000
Box 2008. MS-6335
Oak Ridge. TN 37381-9984
DOE Climate Change and Related Activities
Climate Model Diagnosis and Comparison
The Program for Climate Model Diagnosis and
Intercomparison (PCMDI) assists scientists in
determining the reasons for variation in climate
modeling results. The program compares different
modeling systems by examining the data sets and
parameters used to obtain the results. This service helps
scientists improve or modify their models. PCMDI is
expanding to lend support to the international efforts
being coordinated by the World Climate Research
Program.
Climate Scholarships
An educational initiative of the Global Change Program
is to award competitive fellowships and scholarships at
the postdoctorate. graduate, and undergraduate level.
Special emphasis will be given to involving students in
ongoing research at the National Laboratories to
achieve practical experience in the multidisciplinary
sciences of global change.
For more information please contact:
Oak Ridge Associated Universities
Science / Engineering Education Division
P.O. Box 117
Oak Ridge. TN 37831-0117.
Policy Analysis
The Department conducts analytical studies to examine
current and future emissions of greenhouse gases,
policies for influencing those emissions, impacts from
potential climate change, and policies to adapt to
climate change. In addition to studies, the program also
includes the development of computer models to
estimate greenhouse gas emissions, analysis of the
potential of certain policies to reduce emissions, and the
development of policy-oriented models for estimating
atmospheric effects such as the radiative forcing of
different greenhouse gases.
In early 1989, four congressionally mandated studies
were initiated. These studies are now complete:
• A Compendium of Options for Government Policy to
Encourage Private Sector Responses to Potential
Climate Changes—a study of how private interests can
be encouraged to participate in emissions reductions.
5
1-43
-------
• Global Climate Trends and Greenhouse Gas Data:
Federal Activities in Data Collection. Archiving, and
Dissemination—an analysis of government data
bases, including access and ;:vailability to interested
parties.
• Confronting Climate Change: Strategies for Energy
Research and Development—a study by the National
Academy of Sciences and the National Academy of
Engineering to evaluate the potential of Alternative
Energy Systems to affect greenhouse gas emissions.
• Limiting Greenhouse Gas Emissions in the United
States—an analysis of policy options to achieve
specified levels of emissions reductions in the United
States over the next 10 to 20 years.
Copies of the listed or additional studies can be
obtained by contacting:
U.S. Department of Energy
Office of Environmental Analysis. PE-63
1000 Independence Avenue. SW.
Washington. D.C. 20585.
International Activities
The Department of Energy participates in a number of
international activities in the area of global climate
change:
• The United Nations (U.N.) Intergovernmental
Negotiating Committee (INC) was established in
December 1990 to negotiate an international
framework convention on climate change to be ready
for signature at the U.N. Conference on Environment
and Development to be held in Brazil in June 1992.
• The Intergovernmental Panel on Climate Change
(IPCC) was established in November 1988 under the
joint auspices of the World Meteorological
Organization and the U.N. Environment Program to
assess the science, impacts, and possible responses to
global climate change.
• The Preparatory Committee Meetings for the 1992
U.N. Conference on Environment and Development
(UNCED). The 1992 UNCED is a major environ-
mental activity that is a follow-on conference to the
1972 Stockholm Conference on the Environment. It
will examine strategies on the environment and
development in an attempt to reach specific
agreements and commitments by governments and
international organizations. Virtually all environ-
mental concerns will be addressed, including
biodiversity, transboundarv air pollution, water
quality, and financial and technology transfer issues.
DOE officials represent the Department on U.S.
delegations to these bodies and DOE experts participate
in the preparation of technical assessments. The IPCC
completed its First Assessment Report in August 1990,
which was adopted at Ministerial level at the Second
World Climate Conference in November of that year.
The report is a comprehensive assessment of the state of
knowledge of climate change and a roadmap to the
areas where major uncertainties remain and continuing
research is required. More than 30 DOE scientists
participated in the scientific portion of the report: DOE
experts also contributed heavily to the assessment and
overview of response strategies where the energy sector
was a major area of focus.
The IPCC is undertaking new scientific, technological,
and economic studies in support of the INC
negotiations, and DOE will be a key contributor to
these studies.
The INC has held three negotiating sessions in
February, June, and September 1991. and two additional
sessions are scheduled for December 1991 and
February 1992. Working groups on Commitments and
on Mechanisms have been formed.
These working groups have begun to address the major
issues to be negotiated, including commitments related
to net greenhouse gas emissions: financial and
technological assistance to developing countries:
research needs, monitoring, assessment, and
information exchange; legal and institutional
mechanisms: financial resources: and technological
cooperation. Energy considerations are integral to all
facets of this process.
6
1-44
DOE Climate Change and Related Activities
-------
Part II: Related Activities
Related activities are those that are carried out by DOE
for economic, energy development, and environmental
reasons, yet nonetheless have some bearing on the
matter of climate change.
Scientific Research
DOE conducts fundamental research in physics,
chemistry, biology, medicine, environmental sciences,
ecology, geology, engineering, and mathematics. A
portion of this long-term effort will contribute to our
understanding of the causes and effects of possible
climate change. Atmospheric chemistry, solid earth
studies, and solar radiation physics contribute important
information to support Earth Systems Research.
Ecosystems research also supports Earth Systems
Research efforts. Baseline data have been obtained at
seven DOE national Environmental Research Parks.
DOE arctic sites, and the National Science Foundation 's
Long-Term Ecological Research sites. The objective is
to strengthen the basis for a theoretical understanding of
complex global environmental systems.
Transportation
The U.S. transportation sector accounts for about
27 percent of national energy use. This sector of the
economy is unique in its near total dependence on
petroleum. Research and development (R&D) activities
in DOE's Transportation Program are designed to
improve the efficiency of oil use in the transportation
sector and to increase the availability and use of
alternative fuels. Program efforts center on developing
advanced, high-efficiency alternatives to the internal
combustion engine, evaluating the combustion and
emissions characteristics of alternative fuels,
demonstrating alternative-fuel vehicles in realistic
settings, and developing biofuels from renewable
resources. These activities can reduce the threat of
climate change by developing alternative fuels that emit
less greenhouse gas and by increasing transportation
efficiency to use less fuel for comparable performance.
Alternative Fuels Development
Natural Gas
The United States retains large supplies of natural gas.
some of which are difficult to extract. The Department s
Unconventional Gas Recovery Program seeks to
develop geological data and advanced technologies for
extracting gas from very large but currently
uneconomical gas resources. Estimates of resources and
methods to evaluate potential recovery from deposits
have been improved. Compared to gasoline, natural gas
as a transportation fuel can reduce nitrogen oxide (NOx)
and CO-> emissions, as well as other pollutants.
Biofuels
Biofuels (ethanol and methanol) from biomass (trees,
grasses, waste paper, and so forth) could create a
••closed" C02 system in addition to reducing other
pollutants associated with gasoline. During combustion,
biomass-derived fuels emit COi- However, the trees or
grasses that are harvested to produce the fuel must be
replanted for the next "fuel harvest. During the plants
growth cycle, the COt emitted into the atmosphere
during combustion is reabsorbed by the plants. This
creates a rotating fuel cycle in which there is little or no
net increase in COt emissions.
DOE carries out research to develop processes that
make the mass production of biofuels economically
feasible, as well as works closely with the Department
of Agriculture to develop fast-growing and productive
plants for fuel use (Figure 4).
Vehicle Development
In addition to engine technology development to utilize
alternative fuels, electric battery development is another
component of the Department s vehicle development
program. Electric vehicles could play a key role in
helping to reduce urban air pollution. These vehicles
could also reduce NOv an important greenhouse gas,
and CO-, emissions, depending on the generation
method used to produce the electriciiy 10 charge the
battery (Figure 5).
DOE Climate Change and Related Activities
1-45
7
-------
Figure 4. Sweet Sorghum, Used in the Production of Ethanoi
An electric battery consortium has been formed
bringing together the three major U.S. automakers with
battery developers and electric utilities to cost-share
with DOE a multiyear research program. In addition,
research is being conducted on fuel cells and altemaxive
hybrid propulsion systems. Fuel cells, which use a
chemical process to extract energy from hydrogen, offer
the long-term prospect of efficiency that is twice that of
internal combustion engines, with little or no emissions.
The gas turbine and low-heat rejection diesel are also
being researched for their fuel efficiency and alternative
fuel use potential.
Residential, Commercial,
and Industrial Efficiency
Building, Appliance, and Equipment Efficiency
Residential and commercial buildings account for
36 percent of the Nation 's energy use. Improvement in
the energy efficiency and substitution of renewable
resources to replace carbon-based fuels in this sector
can reduce greenhouse gas emissions. Because of the
size and diversity of the building sector, this task
involves a cooperative effort among government
agencies, industries, manufacturers, professional and
trade associations, and environmental and citizen groups.
In addition to maximizing the efficiency of energy
consumed and the proportion of renewable resources
for the building sector, recently the Department has
focused on finding substitutes to replace building and
appliance materials that contain ozone-depleting
chlorofluorocarbons (CFC's), which are also potent
greenhouse gases. These CFC materials have been used
in foam insulation and refrigerants.
The main elements of a building that affect energy use
are divided into two parts: (1) the envelope, which
includes walls, roof, foundation, doors, and windows;
and (2) building equipment and appliances.
The Envelope
Analysis Tools. The Department supports the
development of computer software to assist builders
and architects in evaluating the efficiency of different
building designs. DOE-2 is a public domain computer
program used for energy analysis of residential and
commercial buildings. It is used to design energy-
efficient new buildings, analyze efficiency improving
modifications for existing buildings, calculate energy
design budgets, and perform cost-benefit analyses of
new technologies (Figure 6).
Window Technology Development Windows, which
are responsible for one-fourth of the energy required to
heat and cool buildings, offer a large opportunity to
save energy. Improving window quality through the use
of treated glass and airtight window framing are key
DOE Climate Change and Related Activities
1-46
-------
Figure 5. Electric Vehicle
factors in the research cosponsored by the Department
and the window industry.
The National Fenestration Rating Council (NFRC), an
organization composed of representatives of the
fenestration industry, building industry. Federal. State
and local governments, utilities, and public interest
groups, has been working to develop a standardized
rating system to provide a measure of the energy
performance of windows, doors, and skylights to be
used for a voluntary rating system.
Passive Solar Design and Construction. Proper
orientation, adding thermal mass, and controlling and
distributing heat allow a building to take advantage of
passive solar energy and reduce the need for energy
expended for heating, cooling, and lighting purposes. In
addition, the use of trees to shade buildings can reduce
energy consumption for cooling.
Appliance and Equipment Efficiency
The Department of Energy will continue its efforts to
improve the energy efficiency of appliances and
equipment. The Department has established standard
test procedures, which are used by manufacturers for
energy-intensive residential appliances (such as
refrigerators, air conditioners, stoves, and furnaces).
HVAC (Heating, Ventilation, and Air Conditioning)
Equipment Space heating and cooling equipment uses
about half of all primary energy consumed in residential
and commercial buildings. Research cosponsored by the
Department and the manufacturing community
concentrates on non-CFC refrigerants, thermally
activated heat pumps, advanced materials and
subsystem components for active solar systems, and
desiccant materials for cooling and dehumidification.
Research for furnaces and boilers focuses on oil-fired
equipment, including controls, fuel atomization.
emissions, venting, and efficiency degradation in
relation to fuel quality.
Lighting. The Department's lighting research efforts
have concentrated on efficiency and the effectiveness of
electric lamps and fixtures. Research has resulted in the
improvement of fluorescent lamps and the development
of electronic ballasts. The Department supported
research that has demonstrated fixture concepts that
enhance efficiency by 15 to 20 percent. A series of
these prototype fixtures is being transferred to industry
for commercialization (Figure 7).
Industrial Process Technologies
Significant cost-effective improvements in industrial
energy efficiency are possible. The Department's
DOE Climate Change and Related Activities
1-47
-------
Figure 6. Energy-Efficient House
industrial efficiency programs concentrate on seven
major areas, common to many industrial operations:
materials processing, electric motor drive, separation
technology, sensors and controls, bioprocessing,
advanced materials, and process heating and cooling.
These programs are expected to yield advanced drying
systems for paper production, improved control systems
for temperature and moisture control, high temperature
heat pumps, improved membrane separation equipment,
new steelmaking processes, and advanced sensors for
critical online process measurements. For the longer
term. DOE is working with industrial1 partners to
Figure 7. Fluorescent Lamp and Heat Sink
10
1-48
develop corrosion-resistant, ultra-high-temperature
materials, low energy sulfur-free pulping processes,
near-net shape casting, and continuous fiber ceramic
composites.
Industrial Waste Minimization
The Department conducts research to improve
industrial energy efficiency. A major emphasis of the
program addresses the fact that a great deal of energy is
wasted each year in the form of embodied energy, of
unused or poorly used raw materials, in the energy
content of industrial waste streams, and in the energy
used to clean up and dispose of wastes. DOE has
recently begun a research program, cost-shared with
industry, to develop technologies to reduce wastes at the
outset rather than cleaning them up after they have been
generated. In addition to reducing waste, this approach
saves money and thus improves the productivity and
competitiveness of U.S. industry.
Electricity Generation and Use
Integrated Resource Planning
Integrated Resource Planning (IRP) is a process by
which utilities decide how to best provide energy-
related services at the least cost to consumers. In the
DOE Climate Change and Related Activities
-------
past, utilities have built new power generating plants to
meet additional demands for energy. A different
approach is to allow utilities the regulatory flexibility,
and financial return needed, to meet customer demand
through efficiency improvement, load management, and
other measures in addition to traditional supply options.
This "stretches" the ability of the same amount of
generated power to meet additional needs, eliminating
the necessity to build additional power plants.
Successful IRP programs can limit future additional
emissions of greenhouse gases by reducing the need for
new fossil fuel powerplants.
The Department provides an information clearinghouse
and technical assistance to States and utilities who are
trying to start or improve their Integrated Resource
Planning efforts. States can share information on and
learn from others how to improve their planning efforts,
as well as keep abreast of regulatory or legislative
changes.
Electricity Generation
Ronewables
A number of renewable energy technologies can
contribute to electricity generation in the future while
helping to reduce greenhouse gas emissions.
Implementation of the National Energy Strategy actions
is projected to increase the amount of electricity
produced from renewable energy by an estimated
16 percent in the year 2010.
Wind Energy. The goal of the Department's wind
research is to improve the efficiency and cost-
competitiveness of wind energy systems to make them
commercially competitive in many areas of the world
(Figure 8). Research activities are focused on solving
the structural fatigue problems of rotors and blades, and
improving manufacturability and reliability. In
cooperation with utilities and system manufacturers, the
program plans to develop a cost-shared advanced wind
turbine development program.
Photovoltaics. The Department's research and
development effons to convert sunlight directly into
electricity (photovoltaics) have assisted in making this
technology economically feasible in remote sites where
other power sources would not be cost-effective. The
program continues to pursue its goal of developing
photovoltaics as an economical bulk power option for
the United States. Utility-sized photovoltaic systems for
sites with high sunlight intensity are expected within
the next few years. DOE-sponsored research is directed
towards increasing conversion efficiencies and reducing
manufacturing costs. DOE will continue to assist the
industry to maintain and extend its world leadership
role in the manufacture and commercial development of
solar equipment, components, and systems.
Solar Thermal. Solar thermal systems use large fields
of mirrors to concentrate the Sun's heat in a working
fluid that is then used to generate electricity. Since
1984. over 355 Megawatts (MW) of solar thermal
electric capacity has been installed in the United States.
DOE-sponsored research is directed towards reducing
costs of large utility systems and producing small cost-
effective nonutility systems.
Geothermal. Geothermal systems generate electricity
using heat from the Earth. Currently, electric generating
capacity using geothermal energy is about 2.800 MW.
Development has been limited to locations where there
is a concentrated heat source. DOE-sponsored research
is directed toward obtaining a better understanding of
geothermal reservoirs and reducing the costs of
exploration.
Hydropower. Hydropower provides over 9 percent of
the Nation's electricity supply with a generating
capacity of over 70,000 MW by harnessing the energy
produced by falling water. DOE effons are directed
toward identifying improved methods of addressing
environmental concerns and removing unnecessary
regulatory barriers to development, especially at
existing dams.
Figure 8. Levelized Cost of Energy
From Wind Turbines
W 5 20
81 82 83 M 85 86 87 88 89 90
YOif
DOE Climate Change and Related Activities
1-49
-------
Nuclear. Nuciear power, which produces no greenhouse
gases, accounts lor aboui 20 percent of the Nation 's
electricity supply. The goal of the nuciear energy
program is to remove undue regulatory and institutional
barriers to the use of safe, economical nuclear power.
New nuclear reactor designs incorporate evolutionary
improvements to current designs. These designs are
ampler. standardized, safer and more economic to build
and operate. DOE will continue to work cioseiv with
industry, utilities, and independent research organi-
zations to make progress in the development of nuclear
power generating technologies.
Fusion. At sufficiently high pressures and temperatures,
(he nuclei of light atoms are forced together, causing the
nuclei to fuse and. in the process, to release energy.
Fusion is a long-ierm energy option that could become
a principal energy source in the next century. It can use
in inexoensive. vast, and secure fuel resource while
offering attractive environmental and safety aspects.
Fusion development aiso drives lechnology advances in
fields as diverse as plasma physics, superconducting
magnets, high-power accelerators and radiofrequency
systems, advanced materials development, and
computer science, to name just a few.
The fusion programs of the world currently have
vigorous international cooperation and collaboration
that include scientific and technological exchanges,
joint experiments, and joint planning activities. The
U.S. program contributes and is a leader in these
activities by emphasizing early international
involvement m planning for major new activities and
facilities.
Fossil
Worldwide burning of fossil fuels is a major source of
greenhouse gas emissions. However, the worldwide
fossil energy resource base is large, and economic
development policies indicate continued reliance on
these fuels.
One option that would reduce the amount of C02
emitted from fossil fuel combustion involves the
utilization of advanced fossil fuel technologies, such as
the clean coal technologies being developed by DOE.
operation. The objective of this program is to achieve
commercial availability of these technologies beginning
in the mid 1990's. Industry cost sharing accounts for
about two-thirds of the total program cost.
CCTs have lower CO; emission rates because of their
higher conversion efficiencies: 40 to 50 percent for
CCTs versus 30 to 35 percent for conventional
technologies with SO-, controls. Table 1 compares COt
emissions from several coal-fired technologies.
Combustion efficiency improvements in countries
currently using fossil energy technologies, together with
increased reliance on clean coal technologies in
developing countries that are projected to have rapidly
increasing emissions of greenhouse gases, couid
achieve reductions in these emissions.
Since so much coal is used worldwide, efficiency
improvements can have a large effect upon reducing
CO; emissions. Figure 9 shows historical global
consumption of coal and other fossil fuels.
Figure 9. C02 Emissions Due to Fossil Fuel
Consumption (1980-1985)
(petagrams carbon)
O
n
-------
Table 1. Carbon Dioxide Emissions from Coal-Fired Technologies
% Change
Efficiency C02/year from conventional
Technology (percent) (million tons) technology
Conventional plant 33
with scrubber
Atmospheric fluidized 36
bed combustion
Pressurized fluidized 40
bed combustion
Gasification combined cycle 42
(with fuel cell) 50
MHD
Policy Analysis
National Energy Strategy
The National Energy Strategy (NES) lays the
foundation for a more efficient, secure, and environ-
mentally sustainable energy future. One chapter of the
NES is devoted to global environmental issues,
primarily global climate change. Consideration of
climate change and other environmental issues is woven
throughout the NES and its action recommendations.
Under current policy scenarios, the global wanning
potential of U.S. greenhouse gas emissions in 2030 is
projected to increase by more than 40 percent over 1990
levels. In contrast, the NES measures, taken together,
are estimated to keep U.S. greenhouse gas emissions, as
measured by global warming potential, at or below
present levels through 2030.
All energy programs described in this inventory were
analyzed by the NES for the effect on greenhouse gas
emissions. Figure 10 shows the NES analysis of the
timeline during which specific energy technologies,
now undergoing R&D, are expected to come into more
widespread use.
Alternative Fuels
To address both energy security and environmental
issues associated with oil use in transportation, a
DOE Climate Change and Related Activities
3.1 0
2.8 -10
2J -19
2.4 -23
2.0 -35
-42
multiyear study of alternative fuels is being conducted.
Methanol, ethanol. compressed natural gas. liquefied
petroleum gases, and electricity are being examined for
the time period 1995-2010.
This study characterizes the emissions of alternative
fuels, relative to gasoline and diesel fuel, for both light
and heavy duty vehicles. Pollutants being examined
include hydrocarbons, nitrogen oxides, carbon monox-
ide, carbon dioxide, particulates, and aldehydes. This
analysis will be used by the Interagency Commission
on Alternative Motor Fuels to make recommenda-
tions for future development of alternative fuels for
transportation.
Conservation and Efficiency Studies
Currently, five ongoing analyses of conservation and
efficiency are underway to examine the structural,
behavioral, technical, and economic factors that explain
trends in energy efficiency and provide estimates of
future potential. The areas being examined are oil
conservation potential, electricity conservation,
transportation fuel efficiency, energy in trade competi-
tiveness. and energy efficiency and the environment.
Additional Studies
The Department also supports policy analysis of
pollutants from energy-related activities, including the
13
55
1.8
1-51
-------
Figure 10. Timeline for Commercial Introduction of Selected Energy Technologies
1990
1995
2000
2005
2010
2016
2020
2025
2030
—
Advanced
Geotnermal
• Binary and
other
advanced
geothemtal
Wmii Materials
Management
• Municipal solid
wests ftu'idized
bad corrmuation
• Dual coating
paint systems
Industrial
• Advanced
systems tor
combustion of
Mack liquor and
tar drying paper
in pulp and paper
mills
• Advanced system
(or refining glass
t t
ANMMtlVI
Liquid Fuels
From Blomaet
• Enzymatic
conversion
• Thermo-
cnemicai
conversion
Fuel Celts
• Autorooiive
applications
• Utility
applications
1
Photovoltaic*
¦- Utttty scale
appfecanofts
X
X
X
4nduatrial
Advanced
process fior
near-net shape
casting of steel
-Advanced cell
design 4or
electrolytic
production of
aluminum
Nuctear
Modular fitgfi
temperature
gas reactor
(MHTQR)
Liquid metal
Sup«rconductMty
• -Higher temperature
superconducting
meterielsin
transportation and
industry
• Magnetic storage
X
X
Automotive Qaa Turbine
Electric Vehicle With
BuHdings
Technology
• Efficient.
non-CFC
refrigerants
• High-efficiency
lighting
• R-10* windows
and elsuio-
- chromic* windows
Electric
Storage
I OCWnOK)Q9OT
• Advanced
batteries
Induetrlel
• Magnetic
Biomoss feedstocks
substitute tor
hydrocarbons
X
Industrial
Advanced
direct
sieelmaking
¦ ¦ — — | - -ia
lil^nVDC LIIIUUOli
applications
X
Advanced Oil Recovery
• Chemical flooding,
miscible flooding, thermal
recovery, polymer
flooding, profile
modification and Infill
drilling
X
Clean Coal Technology
• Atmosphere fhitd&ed bed
• Pressurized fiuidoed bed.
~ combined cycle
• Integrated gasification
combined cycle
Electric
Technologies -
• Advanced wind
turbine
" technology
• Photovonsics
: for distributed
apt**Hnns
• Btamaas
: ffuUoed bed/
• high efficiency
gaeturbinee
dear Coal
Technology
• Advanced
pressurized
¦ huidtzed bad
•-Advenoed
Integrated
gasification
combined
cycle
.£u*ton
DemonatiaUorvPtaat^
(Commercial
Introduction—3046R5-
X
Industrie
• Biological
substitute
thermal
¦ -High lemperatiami
matsnals for host-
engines
J/'
X
-Advanced __
'Fuel CeUe ~
• Solid oxide
-Hydrogen TusWdf^g
X
Oaothennal
¦> Hot dry rockarkfe.
systems
Nucieer
Technology
• Advanced
light wetar
1 Advanced
Diesel Engtne-
14 DOE Climate Change and Related Activities
1-52
-------
development of analytical tools and assessments of
energy technologies. Pollutants of local interest may
interact to influence greenhouse gases or may relate to
climate change in other ways. Policies to control
greenhouse gases may be antagonistic to or synergistic
with control of environmental pollutants.
Related International Activities
The Department of Energy participates in a number of
international activities related to global climate change.
These include bilateral and multilateral discussions on
climate change and energy policy; international R&D
cooperation in science and technology to mitigate
greenhouse gas emissions: and promoting the export of
U.S. technologies and services.
International Agreements
The Department administers approximately 160
international agreements between the Department and
over two dozen countries in Europe, the Americas,
Asia, and Australia. These agreements facilitate
information exchange on pollution control, clean coal
technology, renewable energy, energy efficiency, and
nuclear energy.
Enhanced international collaboration in energy research
and technology development to address environmental
issues such as climate change is an increasingly
important aspect of the Department's international
activities.
For instance, new activities have been launched: a
collaborative S10 million clean coal technology
program with Poland for the retrofit of an existing coal
plant: a joint project with Mexico to help monitor air
quality in Mexico City and identify mitigating
measures; and a joint study with the European
Community to study the full fuel cycle costs of various
energy systems.
Bilateral and Multilateral Cooperation
DOE conducts bilateral consultations on energy issues
to provide an opportunity for international energy
information exchange and for promotion of U.S. views
on energy-related environmental concerns. Bilateral
consultations are held on a periodic basis with Canada,
Indonesia, Japan, Norway, Mexico, the Republic of
Korea. Venezuela, and several oil-producing countries
in the Middle East. DOE is also represented in
multilateral programs dealing with energy and
environmental issues. These include the international
Energy Agency (IEA). the International Atomic Energy
Agency, the United Nations Economic Commission for
Europe, and the Minerals and Energy Forum of the
Pacific Economic Cooperation Conference.
Environmental issues such as climate change are now a
major feature of the agendas and discussions of these
organizations. For instance, the IEA has completed a
number of major studies on the environment, has
surveyed and compiled an inventory of the climate
policies of its member nations, and is undertaking new
studies on the technological potential, costs, and energy
market impacts of responses to climate change.
Through its participation in the various IEA Standing
Groups, in the IEA Governing Board, and at Ministerial
meetings of the IEA, the Department has been
instrumental in shaping this agenda.
Promoting U.S. Exports
The Department participates in interagency efforts to
promote the export of U.S. energy-related technology
and services. DOE is currently implementing a major
new initiative to identify and promote the export of U.S.
oil and gas technology and services, conservation and
renewable technologies, clean coal technology, and
nuclear energy technology. These technologies can
promote environmental quality by replacing outmoded
or less efficient technologies, encouraging enhanced
energy efficiency, and promoting cleaner use of fossil
fueis and safer use of nuclear energy.
DOE Climate Change and Related Activities
1-53
15
-------
Working Group of the DOE Climate Change Executive Committee
Scientific Research
Policy, Planning and Analysis
Conservation and Renewables
Fossil Energy
Nuclear Energy
International Affairs
Environment. Safety, and Health
Office of the Secretary
General Counsel
Public Affairs
Congressional Programs
Anstides Patrinos. ER-76
George Jordy, ER-30
Howard Gruenspecht. PE-6
Ted Williams. PE-63
Kenneth Freidman. CE-20
Paul Baily, FE-1
Bob Kane. FE^
Richard Oehl, NE-44
Richie Williamson. IE-10
Kathleen Rees, IE-12
Denise Dwyer. IE-141
Yvonne Weber. EH-252
Rick Bradley, S-I
Mike Delello, S-l
Rebecca Thomson, GC-15
Phil Keif. PA-3
Carolyn Gay, CP-1
William Brennan, CP-33
301-903-4375
301-903-2971
202-586-5316
202-586-2061
202-586-9236
202-586-6660
202-586-4753
301-903-2948
202-586-5493
202-586-5902
202-586-6384
202-586-7598
202-596-8008
202-586-0509
202-586-0393
202-586-5806
202-586-5450
202-586-4827
U.S. Department of Energy
1000 Independence Avenue. SW.
Washington. D.C. 20585
The work described in this paper was not funded by the U.S. Environmental Protection Agency. The
contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred.
DOE Climate Change and Related Activities 17
, _ v54 _ ¦ —
i.
-------
1-F
EPRI'S GREENHOUSE GAS EMISSIONS ASSESSMENT
AND MANAGEMENT RESEARCH PROGRAM
D. F. Spencer, G. M. Hidy
Electric Power Research Institute
Palo Alto, CA 94304
ABSTRACT
This paper will briefly review the rationale and need for an electric utility Greenhouse Gas
(GHG) emissions assessment and management research program, as well as potential cost
implications to society of short-term control requirements. The balance of the paper will focus on
EPRI's directly-related R&D program and its key elements, namely a) model development and
model evaluation activities, b) ecological effects, c) management/mitigation research, and d) the
development of an integrated assessment framework. EPRI's dixecdy-related GHG research
program, including cofunding, is expected to expend approximately $60.0 million over the next
four years to address key aspects of this significant environmental issue.
BACKGROUND AND INTRODUCTION
Over the last 100 years, the dramatic enhancement of national economies and individual
well-being, particularly in industrialized nations, has resulted in large increases in the use of energy
and, particularly, fossil fuels. Simultaneously, large land areas in both developed and developing
countries have been convened from forests and grasslands to agricultural lands, with concomitant
impact on the global carbon balance. Finally, the world population explosion, as well as
lengthening life spans, have propelled the world to levels of activity which are challenging the
environment locally, regionily, and globally.
These activities arc producing a wide variety of gases, which when released to the
atmosphere, produce a significant biogeochemical perturbation of the natural environmental
system. The gases include carbon dioxide, sulphur oxides, nitrous oxides, chlorofluorocarbons,
methane, volatile organic compounds, etc. In addition, the transitioning of forested lands to
agricultural uses alters a) the biofixation potential of specific regions, b) soil nutrient and moisture
contents, c) surface albedo, d) ecological habitats, etc. These perturbations of the earth's
atmospheric composition and land use, in turn, result in modified physical, chemical and biological
properties of the atmosphere, land, and ocean systems.
Two of these modified properties have been the focus for many studies of the impacts of
so-called "greenhouse gases," namely the projected mean temperature increase of the earth's
atmosphere and sea-level rise. Obviously, many other properties will also be modulated
1-55
-------
simultaneously including those arising as a result of induced changes resulting from temperature
change, etc. Our present ability to assess most first and second order effects and system feedback,
and to properly consider all system responses is gready limited.
Examples of limitations include our present inability to a) adequately account for changes in
solar radiation fluxes, b) properly evaluate the role of cloud feedback on climate, c) properly
consider aerosol cooling, from natural and anthropogenic sources, d) include biofeedback, both
terrestrial and oceanic, etc. Each of these processes plays an important role in maintaining either
our earth's radiative or chemical balance. It is incumbent upon us to develop a more thorough
understanding of these processes, and associated effects, in order to develop meaningful policy
and technological responses.
It is the purpose of EPRJ's Greenhouse Gas (GHG) Emissions research and development
(R&D) program to a) assess the overall scientific level of understanding of these limitations and
identify research opportunities, b) assess the impact of the U.S. electric utility industry on GHG
emissions and the effects of climate change on the industry, c) estimate expected effects of
increasing concentrations of GHG and economic consequences, domestically and internationally,
and d) develop management and mitigation options to minimize deleterious effects on the climate,
ecological and human systems.
PERSPECTIVE ON U.S. ELECTRIC UTILITY
GREENHOUSE GAS EMISSIONS
At present, the U.S. electric power industry generates approximately 33% of the carbon
dioxide produced in the U.S.(l) This is equivalent to 7% of the total anthropogenic CO2 generated
annually on a global basis and accounts for less than 4% of the global warming potential (GWP).
The GWP provides a means for a simple comparison of the radiative forcing of various greenhouse
gases. It is, by definition, "the time integrated commitment to climate forcing from the
instantaneous release of 1 kgm of trace gas expressed relative to 1 kgm of carbon dioxide."
The U.S. power industry has traditionally depended on coal as its preferred fuel. In 1991,
the U.S. power industry consumed more than 800 million tons of coal to generate more than half
of the electricity produced in the U.S. Present coal-fired plants have the reputation of being "dirty"
and environmentally acceptable, only with retrofit emission control technology.. Trends in new
coal-fired generation, spurred by the Clean Air Act and its amendments, are leading to a whole new
breed of higher efficiency, environmentally benign systems.(2)
Although short-term generation additions in the U.S. are projected to primarily rely on
natural gas, it is expected that coal will play an important and perhaps increasing role in future
U.S. electricity generation well into the mid 21st century.' As a result, it is expected that overall
emissions of carbon dioxide from powerplants will rise with increased demand for electricity over
the next 15 to 20 years.
On the other hand, projected increases in electricity end use efficiency and substitution of
electricity for other end use fuels; e.g., natural gas, can lead to substantial overall reductions in
CO2 emissions. These so-called "wiser and wider" uses of electricity are forecast to have the
potential to reduce U.S. C02 emissions by nearly 700 million tons by 2010, or approximately
13% of our nation's present CO2 emissions.(3) Therefore, it is imperative that we evaluate the
benefits of electric power from the generating plant through the final end use to properly assess the
contribution to GHG emission reductions.
1-56
-------
OPTIONS FOR GREENHOUSE GAS EMISSIONS MANAGEMENT
Over the short-term (the next five to ten years), the options available to the electric power
industry for reducing GHG emissions are limited. Any such controls, at the powerplant site,
would have to be made on the existing 494,000 Mwe of fossil generating capacity to significantly
reduce emissions. Most of these units do not have space for so-called CO2 scrubbers, and cost
and efficiency losses from such systems would make them totally uneconomical.(4)
The only real short-term options are a) improving end-use efficiency, b) improving the
efficiency of existing fossil generating units, c) carbon dioxide systems dispatch, and d) natural
gas substitution. Each of these options is discussed in reference 1, with the following estimated
reduction potentials over the next ten years:
• It is estimated that between 3% and 8% of the unconstrained CO2 powerplant
emissions in 2000 could be eliminated by actively promoting end-use efficiency
improvements.
Although heat rate efficiency gains of 1 % to 5% are possible with existing
fossil units, new Clean Air Act amendments are likely to result in a net decrease
in the efficiency of coal-fired units.
If gas/oil-fired units were dispatched in place of coal, utilities might further reduce
CO2 emissions by 3% to 5%; however, this would obviously produce higher
cost power.
The greatest CO2 emission reduction potential would result from mass
substitution of natural gas for coal. If CO2 emission levels were to be held
to today's levels, nine to ten trillion cubic feet of new natural gas annually would
be used by the power industry. This is extremely unlikely and extremely
costly. Cost estimates, taken from reference 1, indicate that maintaining
constant CO2 emissions over the next decade would require annual expenditures
of $17 to $42 billion ($1990) in 2000. In addition, substantial new gas pipeline
capacity would be required and it is likely that the additional demand for natural
gas would result in even higher gas prices; i.e., greater annual costs than the
estimated $17 to $42 billion.
In the longer term (beyond 2000), many new options will become available to the electric
power industry and its customers to effect significant reductions in GHG emissions. These
options (5) include further improvements in end-use efficiency, powerplant generation efficiency
improvement, substitution of nuclear and renewable powerplants for fossil power, carbon dioxide
scrubbing and sequestration, and terrestrial and, perhaps, oceanic phytomass production and
storage or substitution for fossil fuels. Further, there are opportunities for selective natural gas
substitution or co-firing, for development of integrated energy facilities, and the transfer of energy-
efficient electric technologies from developed countries to developing countries.
Of course, all of these control strategies come at some price, and costs of carbon emissions
reduction programs range from net benefits of end-use efficiency gains to costs of $350 to $700
per ton of CQ2 sequestered or averted.(5) A long-term management, mitigation strategy, based on
minimizing the costs to society, is a major objective of EPRI's assessment and GHG emissions
reduction research program.
1-57
-------
EPRI'S GREENHOUSE GAS RESEARCH PROGRAM
EPRI has been supporting greenhouse gas research for over a decade. Early sponsored
research focused on carbon cycle modeling, the role of the ocean in CO2 uptake, and two
international symposia on "Carbon Dioxide Transfer in the Atmosphere-Ocean-Terrestrial System."
This early work included support for Dr. Charles Keeling's key geophysical monograph series
published in 1989.(6) This work included:
Three-dimensional modeling of atmospheric CO2 transport, including the influence
of El Nino events on biospheric and oceanic CO2 exchange and identification of the
biosphere becoming a key CO2 source term in the mid-1970s time frame.
Modeling of CO2 sources and sinks including fossil fuel CO2, exchange between
the terrestrial biosphere and atmosphere, and air-sea exchange, driven by prescribed
temporal and spatial variations of CO2 in surface waters.
Modeling of the mean annual cycle and interannual variations in atmospheric C02,
as well as comparison with measurements over a 25-year period from pole to pole
over the Pacific Ocean.
Other research included development of an 11 compartment (box) model of the ocean-
atmosphere system which predicted large increases in Antarctic surface water biological production
during the last ice age. This may have resulted in a net organic flux into the oceans, lowering the
atmospheric CO2 concentration by 90 to 110 ppm.(7) The workshops held in 1985 and 1988,
brought together international experts in the global carbon cycle to discuss the state of carbon cycle
modeling, ice core and tree ring data, ocean circulation, terrestrial biospheric changes, etc. These
workshops played a key role in facilitating information exchange among experts on various
elements of the carbon cycle.
In the last few years, EPRI's directly-related GHG research and development program has
greatiy expanded to include a) the reliability and completeness of climate prediction studies, b)
assessment of effects on human and natural systems, c) GHG management/mitigation options, and
d) a focus on the development of an integrated assessment framework. These activities are in
addition to a broad range of indirect efforts which would support electric utility CO2 emission
reductions including a) increased emphasis on end-use efficiency improvements, b) advanced,
higher efficiency, fossil power cycles, c) renewables and biomass technologies, and d) advanced
nuclear powerplants.
EPRI's research priorities have been established to
• Assess uncertainties associated with General Circulation Model process and data
limitations,
• Provide bounding analyses of physical, ecological, and economic impacts resulting
from global climate change,
Identify and develop "least cost" management options for the power industry,
including adaptation and mitigation.
1-58
-------
ASSESSMENT OF CLIMATE MODELS
In 1991, EPRI, in conjunction with the National Science Foundation, and a group of
international organizations, initiated a program to assess the accuracy and uncertainties associated
with climaie models. This program is known as the Model Evaluation Consortium for Climate
Assessment (MECCA). The objectives of the MECCA program are to: a) quantify the probably
range of future climate change, b) provide policy-makers with information that could be used to
coordinate decisions with scientific developments, and c) identify key topics needing research to
improve climate forecasts.
MECCA'S research plan (8) includes two key elements, a set of carefully selected model
experiments, designed to supplement ongoing work by other modeling groups and an analysis
strategy that focuses on the impact-oriented needs of policy-makers and the private sector. The
primary elements of the modeling strategy include:
• Large-scale climate changes including global temperature, precipitation, and
pressure patterns,
Regional climate simulation, at subcontinental grid scales to aid an understanding of
local impacts on agriculture, commercial, and industrial activities,
• Extremum events that might be associated with climate change such
as droughts, floods, storms, prevailing lows, etc., and
• Model evaluation experiments which simulate today's climate in order to identify
any potential systematic modeling errors and needs for model improvement.
Current MECCA experiments include evaluations of the sensitivity of climate model
outputs to a) CO2 concentration, b) model grid size, c) atmosphere-ocean coupling, d) land surface
interactions; e.g., tropical deforestation, e) ice dynamics, and f) cloud parameterization. In
addition, regional climate studies and general circulation model prediction variability in century-
long simulations are being studied.
An example of the evaluations underway is an estimate of the global mean equilibrium
atmospheric temperature as a function of ambient CO2 concentration. These studies have shown
that the mean global atmospheric temperature rises very non-linearly with CO2 concentration.
Maximum equilibrium temperatures of slightly less than 300"K are predicted for CO2 partial
pressures of 1000 ppm. The estimated temperature rise for a doubling of CO2 from 280 ppm to
560 ppm is approximately 5*K, consistent with other model projections.
Another of the studies underway is evaluating the relative influence of the sensitivity of
surface temperature to shortwave (solar constant) changes as compared with longwave (CO2)
changes. Preliminary results indicate a nearly linear response of temperature to variations of the
solar constant of + 5%, as compared with the logarithmic response noted above for changes in
CO2 concentration. Temperature sensitivity about the present value of the solar constant is
approximately 1"K per 1% change in the solar constant.
A third analysis is focusing on the use of different surface hydrology models in General
Circulation Models (GCMs). The models compare results from so-called "bucket" models with
more accurate variable infiltration capacity (VIC) models-. Results show that evaporation rates in
the two models are very similar over a GCM grid cell; however, the bucket model predicts much
1-59
-------
higher near-surface soil moisture contents. These are obviously very important in estimating
biological growth rates and carbon fixation potentials.
A fourth study is evaluating the role of cirrus cloud albedo on global temperature. The
suggestion is that penetrative convection under the condition of warm ocean temperatures is strong
enough to drive large amounts of moisture into the troposphere and to generate dense cirrus clouds
capable of substantially increasing the earth's albedo. Preliminary results indicate that when the
ocean surface temperature exceeds 303*K, the middle and high altitude cloud albedo rises rapidly
to levels comparable to low-level clouds (-0.6). This increase in cirrus cloud albedo keeps the
tropica] temperature of the simple mixed layer ocean near 303*K and cools the entire planet Thus,
the overall increase in planetary albedo is creating a negative feedback which limits surface air
temperature increases.
Another key element of EPRI's GHG modeling efforts has been the development of the
personal computer based global carbon cycle model, GLOCO.(9) The GLOCO model integrates a
set of mechanistic modules of the atmosphere, oceans, and terrestrial ecosystems into a single
global carbon cycle model. In additional to representing the basic reactions of the global carbon
cycle, the model includes the cycling of nutrients that limit the growth of marine and terrestrial
biota.
The terrestrial ecosystems are represented by six biomes - tropical, temperate, and boreal
forests, grassland, tundra, and desert. Carbon in each biome in the model occurs as plant and soil
matter. The primary process driving the terrestrial carbon cycle is the fixation of atmospheric
carbon dioxide as organic carbon via photosynthesis. The earth's atmosphere is represented as a
single box in the model, with two atmospheric carbon compounds, namely C02 and CH4. The
concentrations of each are calculated from the balance of inputs to the atmosphere, uptake by the
oceans and terrestrial ecosystems, and atmospheric reactions.
The oceans model includes the biogeochemical cycling of five chemical components,
namely total inorganic carbon, dissolved organic carbon, particulate organic carbon, alkalinity, and
phosphate. Both physical and biological carbon exchange between the atmosphere and ocean ait
included in the model.
GLOCO was calibrated by assuming that only fossil fuel carbon and land-use changes
altered the atmospheric CO2 concentration from 1700 to present. The calculated CO2
concentrations are within 5 ppm of ice-core inferred values for the period 1700-1950 and within 2
ppm of the Mauna Loa measured values (1958-1990).
GLOCO projects future atmospheric carbon dioxide concentrations to the year 2100
utilizing IPCC fossil fuel emission scenarios and projected land-use conversions. The model is
particularly capable of permitting a number of sensitivity tests to examine the implications of
uncertainties in key parameters or permitting "worst" case analyses to be performed Examples of
variations in input parameters which have been assessed are:
• Simultaneous carbon and nitrogen fertilization effects on forest ecosystems,
• Evaluation of the effect of different C02 gas exchange fluxes (atmosphere/
ocean) on the oceanic uptake of CO2,
• Impacts of increased oceanic upwelling velocity on CO2 uptake.
Assessments of the sensitivity of atmospheric carbon dioxide to oceanic
productivity.
1-60
-------
In genera], GLOCO offers a means to assess in a globally aggregated way, the relative importance
of a variety of physical and biological processes on future concentrations of atmospheric carbon
dioxide.
ECOLOGICAL EFFECTS AND CO2 MANAGEMENT
EPRI's GHG ecological effects program focuses on development of tools for conducting
assessments and experimental programs to evaluate forest response to C02- The potential for
ecological effects is one of the primary concerns expressed by environmentalists. Preliminary
analyses suggest that unmanaged ecosystems may be the most sensitive of the potential impact
areas, and the levels of atmospheric carbon dioxide and rate of change of mean atmospheric
temperature are extremely important parameters. Of course, it is very important to know whether
the severity of effects accelerates rapidly with a 2*C or 4*C temperature increase, or at a rate of
change of temperature of 0.1 *C per decade or 0.2"C per decade. Another important element may
be whether these increases reflect average maximum daylight temperatures or average minimum
nighttime temperatures, as is indicated by some models.
EPRI research is exploring the range of possible effects of climate change, develop models
and data for assessing climate change effects and quantifying uncertainties, and develop methods
for utilities to use in evaluating and managing power system related water resources susceptible to
climate induced changes.
Two of our present GHG ecological effects studies focus on biomass production in loblolly
pines at elevated CO2 partial pressures and the effects of elevated CO2 on grasslands. Results of
initial growth studies of loblolly pine indicate that elevated C02 alone will have little influence on
biomass growth; however, addition of other nutrients, specifically nitrogen and phosphorus, can
increase biomass yields up to 25% over yields with today's atmospheric CO2 concentrations.
Another study being performed for EPRI by the Nature Conservancy is evaluating the
effects of climate change on plant diversity in North America. The impacts of climate change on
14,000 plant species is being assessed, including stand geographical migration patterns and species
losses.
EPRI is also assessing a variety of indirect and direct management mitigation options to
limit ecological or other damage. There presently is little agreement about which potential impacts
are of greatest significance or concern; and generally accepted methods for cost/benefit analyses for
slowing climate alteration do not exist. Further, benefits may accrue to certain world regions while
costs may be borne by another region.
EPRI's primary research emphasis will be to develop "least cost" systems to ameliorate
GHG emissions to acceptable levels. Of course, the degree to which each component of such a
strategy can be utilized to mitigate the problem is dependent, not only on the total control level to be
achieved, but also the degree to which each alternative actually penetrates the market as a function
of Oilie.
Reference 5 reviews the variety of options available to reduce C02 emissions and
prioritizes them on the basis of cost per ton of C averted or sequestered. It is clear that energy
efficiency improvements, both at the end-use and at the power production level, should be
emphasized in addressing the GHG control issue from an electric utility viewpoint Beyond these
efficiency measures, there are a range of options for limiting CO2 emissions; e.g., nuclear and
renewables that are more attractive than the "C02 scrubbing and sequestration" approach.
1-61
-------
EPRI is sponsoring research relating to range of biospheric options for mitigating
atmospheric carbon dioxide levels.(lO) Terrestrial approaches include halophyte production in
semi-arid regions and short-rotation woody crop production and utilization. Ocean marine systems
research includes a) evaluating the CC>2 sequestration capability of the oceans and sediments, b)
ocean circulation studies, c) sequestering carbon dioxide in the form of clathrates in the deep
ocean, d) development of free-floating or structured macroalgal species/systems to biologically fix
C02, and e) optimal conversion/sequestration strategies.
All of these efforts focus on increasing yields per unit area of biomass, assessing long-term
storage potential, and estimating sequestered (or averted) carbon costs per ton. A combination of
these processes could reduce atmospheric loading by 2.0 to 10.0 gigatons per year at costs of $100
to $300 per ton of carbon. Although none of the biological approaches offers a clear cost
advantage over CO2 scrubbing, they may offer a preferred solution since infrastructure, transport,
etc., requirements may be less cumbersome. In addition, there may be additional advantages
associated with energy or food byproducts which could alter the relative benefit of biospheric
systems.
In addition to our CO2 mitigation program, EPR] is aggressively contributing to the
development of non-ozone depleting; i.e., non-halogenated hydrocarbon refrigerants, which will
also have a minimum impact on the global radiative balance. A major program has just been
initiated to develop new working fluids for unitary heat pumps, air conditioners, and electric
chillers which use neither chlorofluorocarbons (CFCs) or hydrochlorofluorocarbons (HCFCs).
HCFC and CFC based vapor compression systems account for 23% of utility loads and revenues;
thus developing substitute working fluids is a high priority. The two parallel projects are aimed at
replacing HCFC 22 in heat pump and small building cooling applications and HCFC 123 for large
system cooling. The challenge will be to develop substitute working fluids with no decrease in the
cycle efficiency. Pure fluids, azeotropic and non-azeotropic mixtures will be characterized and
evaluated.
INTEGRATED ASSESSMENT FRAMEWORK
The final major component of EPRI's Greenhouse Gas Emissions Assessment and
Management Research Program is the development of an integrated assessment framework. The
purpose of this research program is to provide a consistent and comprehensive approach for
evaluation of GHG management proposals and associated R&D strategies. The strategy is being
designed to provide an evaluation framework for measuring the impact of GHG control strategies
on the overall domestic, as well as international, economies, major industries, and the public.
In concert with the National Science Foundation and the Department of Energy, EPRI is
supporting the construction of a group of models which will synthesize information on the
relationship between human activities and GHG emissions, the effect of GHG emissions on
climate, and the impact of climate changes on human and natural systems. By linking this
information together in a cohesive way, analysts will then be able to evaluate the costs and benefits
of policy proposals in a consistent manner, as well as evaluate the potential value of alternative
R&D strategies. The results from a first analysis using the framework should be available by 1994
and periodic refinements will be made throughout the decade as new knowledge and methods
evolve.
The framework will consist of a series of linked modules representing the major processes,
including a) the emissions of radiatively important gases, b) natural system disposition and
1-62
-------
reactions to the emissions of these gases, and c) the reaction of human and natural systems to
changes in the atmosphere-climate system. As we all know, the development of such an
assessment framework, with its scope and complexity, is a formidable task. Initially, three
contractors will be developing alternative approaches. This collection of models will include
estimates of "damage functions" to develop a specific quantitative impact and risk assessment
basis, both temporally and regionally.
Our approach to the development of an integrated assessment framework must be iterative.
At the present time, our understanding of each of the relevant processes and how they interact is
incomplete. However, enough is known to permit initial development of the entire system. At that
point, the framework can be used to identify those elements of the climate problem that contribute
the largest uncertainties to our ability to provide knowledge or predict outcomes of importance to
policy-makers. For example, we will be able to explore the sensitivity of policy proposals to
different assumptions about natural system processes. In this way, we can identify critical
uncertainties and identify where additional research is most needed. The framework can thus be
used to help guide its own future development.
Another important component of EPRI's risk assessment activities includes evaluations of
the impact of climate changes on utility systems and their operations, specifically, the implications
of potential global, or regional warming on air-conditioning loads, cooling water availability, and
potential implications to hydroelectric generating capacity. To date, our assessments indicate that
expected impacts on utility systems operations are minimal; however, as we develop better
estimates of potential regional climatic changes, there may be regions wherein these impacts are
important.
EPRI'S GHG R&D RESEARCH STRATEGY AND PROGRAM EMPHASIS
As indicated above, a focal point of EPRI's research on climate change is the development
of an integrated assessment framework that will be used to compare the costs and benefits of
alternative climate change management proposals. An important feature of the integrated
assessment framework is the ability to treat uncertainty probabilistically rather than generating
single point estimates of expected outcomes for given scenarios.
Given the vast number of scientific, political, economic, and technological uncertainties
associated with global climate change concerns, the possibilities for designing a research program
are myriad. In recognition of these factors, as well as our awareness that the U.S. federal
government will be spending approximately $1.0 billion annually, the remainder of the EPRI
program has been designed based on meeting one or more of the following criteria:
The research should provide input to integrated assessments in the form of
models and estimates of key parameters and their associated uncertainty.
The research should address key electric utility issues.
• The research should help identify and fill critical gaps in the research being
done by the federal government and other organizations through the world.
• The research should help influence key elements of the national program; thus
achieving a significant leveraging of EPRI's funds.
1-63
-------
Through a carefully designed program, carved out in a timely manner, EPRI research can help
decision-makers in the electric utility industry and elsewhere understand the implications of
proposals for addressing the issue of climate change.
As was stated previously, EPRI recently has significantly expanded its GHG R&D
program. The Institute is planning to expend approximately $40 million on direct global change
research over the next four years, with annual funding by 1995 of $12 to $15 million. In addition,
a number of programs are being cofunded, with estimated cofunding of $20-$25 million over the
period 1992-1995. The allocation of these R&D funds over the next four years are expected to be:
• Integrated Assessment Framework
• Glob*U Carbon Cycle
• Climate Change Prediction
• Effects
• Valuation of Effects
• Costs
• Mitigation and Adaptation
Total (92-95)
The work described in this paper was not funded by the
coMents do not necessarily reflec, me views of the A«ency and no official endorsement should"be i^'
$ 4.4
3.2
23.4
9.7
4.5
13.4
1.4
$60.0
U.S. Environmental Protection Agency. The
1-64
-------
REFERENCES
1. Giuckman, M. J., et al. Global Climate Change Current Issues, Electric Power Research
Institute Report, December 1991.
2. Wolk, R. H., et al, Advanced Coal Systems for Power Generation, Proceedings of the IEA
Conference on Technology Responses to Global Environmental Challenges, Kyoto, Japan,
November 6-8, 1991.
3. Fickett, A. P., Meeting the Needs of Our Customers, Presentation at Edison Electric
Institute 59th Annual Convention, San Diego, California, 5 June, 1991.
4. Engineering and Economic Evaluation of C02 Removal From Fossil-Fuel-Fired Power
Plants, EPRI Report EE-7365, Vol. 1, Fluor Daniel, Inc., Irvine, California, June 1991.
5. Spencer, D. F., A Preliminary Assessment of Carbon Dioxide Mitigation Options, Annual
Reviews of Energy and the Environment, 16: 259-73, 1991.
6. Keeling, C.D., et al, A Three Dimensional Model of Atmospheric CO2 Transport Based on
Observed Winds, Four-Pan Series in Geophysical Monograph 55, American Geophysical
Union, 1989 (David Peterson, editor).
7. Keir, R. S., On the Late Pleistocene Ocean Geochemistry and Circulation,
Paleaoceanography, Vol. 3, No. 4, pp. 413-445, August 1988.
8. Model Evaluation Consortium for Climate Assessment, Consortium Brochure, 1991.
9. Gherini, S., Modeling the Global Carbon Cycle: Implications of Simultaneous Carbon
and Nutrient Fertilization, April 21, 1992.
10. Alpert, S. B., Spencer, D. F., and Hidy, G., Options for Mitigating Atmospheric Carbon
Dioxide Levels, Proceedings of the IEA First International Conference on Carbon Dioxide
Removal, Amsterdam, Netherlands, March 4-6, 1992.
1-65
-------
Paper1-G
GLOBAL EMISSIONS DATABASE (GloED) SOFTWARE
by: Lee L. Beck
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
ABSTRACT
The EPA Office of Research and Development has developed a powerful software
package called the Global Emissions Database (GloED). GloED is a user-friendly, menu-
driven tool for storage and retrieval of emissions factors and activity data on a
country-specific basis. Data can be selected from databases resident within GloED
and/or imputted by the user. The data are used to construct emissions scenarios lor
the countries and sources selected. References are linked to the data to ensure dear
data pedigree. The scenario outputs can be displayed on thematic global maps or
other graphic outputs such as bar or pie charts, in addition, data files can be
exported as Lotus 1-2-3, dBase, or ASCII files, and graphics can be saved as a .PCX
file or exported to a printer. This paper describes GloED and how it works. It also
presents future plans for software enhancements and populating the databases.
1-66
-------
BACKGROUND AND INTRODUCTION
The Global Emissions Database (GloED) was designed initially as an internal
data management tool to handle the large number of greenhouse gas data generated
as the result of international greenhouse gas research activity. Not only was there a
large amount of data but the databases were rapidly changing based on continuing
studies and new information. Initial attempts at handling the data electronically
Involved using commercial software such as Lotus 1-2-3 and dBase. These initial
efforts were frustrating partially because of the limitations of the software and
partially because of the limitations of user expertise. Initial limitations included keeping
track of which data sets were used in constructing emissions inventories and
identifying the quality of the data in each data set. It soon became apparent that a
better system needed to be developed by professionals in software development.
The GloED software subsequently developed is personal computer (PC)-based
and very user-friendly with little or no computer expertise needed. Using GloED, the
storage and retrieval of data is quick and easy. This is important so that updates
can be stored as better data become available. Consequently, current best
estimates are always available. Another advantage of GloED is that units specified by
the user are automatically generated by GloED. If the user calls for units that are not
recognized by GloED, then GloED allows for the input of an algorithm that will define
the new unit.
Another important attribute of GloED is that references are linked to the data.
Consequently, there is the ability of the user to establish the origin of every individual
piece of data in the scenario constructed. An additional advantage of GloED is its
ability to interface with other software packages such as Lotus 1-2-3 and dBase for
those scientists and engineers familiar with these commercial software packages.
After establishing the utility of GloED as an internal data handling tool, it was
presented to an emissions workshop sponsored by the Intergovernmental Panel for
Climate Change (IPCC) In December 19911. As a result of this exposure to the IPCC
and the Organisation for Economic Co-operation and Development (OECD), interest
grew in GloED as a standardized tool which could be used by all researchers to
develop quality assured, country-specific, emissions inventories. The United Nations
Conference on Environment and Development (UNCED) in Rio de Janeiro just 2 months
ago underscored the need for a system to establish baseline emissions and to track
emissions reduction progress by country. GloED has the potential for providing this
system for implementing of the goals proposed at this historic Earth Summit meeting.
GLOED DESCRIPTION
GloED is a software system designed as a tool for generating estimates of
global emissions. GloED generates emission inventories by combining information
about activities with pollutant-specific emission factors for those activities. Activities
1-67
-------
are defined in terms of processes that occur in a specific pollutant source category In
a specific country or at a specific latitude and longitude. Activities are grouped into
discrete data sets within the GIoED system. The user selects one or more data sets
and then has the option of narrowing the scope of the inventory by selecting a limited
number of countries, source categories, and pollutants. The final set of data selected
is called a scenario. GIoED also can accept data provided by user-input.
Consequently, the emissions inventories can be updated as new data become available
to the user. GioED calculates an emissions inventory based on the scenario generated
by the user and can present summaries of the inventory graphically and in textual
form.
The contents of the emissions inventories can be reported in a variety of ways.
A text summary of the emissions inventory will print a tabular breakdown of the
results by country, source category, and/or pollutant. GIoED can develop a pie chart
or bar chart showing the top pollutants or cpuntries or source categories in a form
that allows easy comparison among them. Finally, GIoED can project the results of an
emissions inventory onto a global map, using different colors to designate the type
and distribution of pollutants in the selected scenario. All of these output formats can
be viewed on the screen, saved to a file, or printed as a hard copy. The data can
also be exported to Lotus, dBase, or ASCII.
USING GLOED
Each level of the program has a menu that allows the user to select the
operations that the program should perform at that level. The user can select the
actions in the menu either by clicking a mouse on the desired menu selection or by
typing the first letter of that selection. The GIoED main menu always appears along
the top of the screen and is a set of pull-down menus, which means that the user can
"pull down" further options by selecting a menu item. When the user selects a menu
option-either with a mouse or with the cursor keys-GloED will lead the user to the
screens that apply to that menu option.
Most of the menu items selected will call up a screen of 'scroll boxes* that
allow the user to define more specifically the way in which the selected menu function
is performed. To move among the different scroll boxes, the mouse is used to click in
the desired box, or the user can press the [TAB] key to move dockwise-or [SHIFT]
and [TAB] to move counterclockwise-through the scroll boxes and buttons on the
screen. The user can move within the scroll boxes either by using the cursor (or
arrow) keys on the keyboard or by clicking with the mouse on the "scroll bar" (the
vertical shaded strip with arrows at top and bottom) on the right-hand side of each
scroll box. The user can jump quickly to the very first entry in the scroll box by
pressing the [HOME] key on the keyboard and can jump to the last entry in the scroll
box by pressing the [END] key on the keyboard. The user can view the next or
previous boxful of information in the scroll box by pressing [PAGE UP] or [PAGE
1-68
-------
DOWN], respectively. When the item to be selected is highlighted, the user can select
it by pressing [ENTER] or by clicking on it with the mouse.
Any of three methods can be used to exit one of the screens in GloED at any
time:
(1) Pressing the "escape" [ESC] key;
(2) Moving the mouse to the on-screen [CANCEL] button, and then "clicking*
the button on the mouse to select cancel; or
(3) Using the [TAB] key to tab clockwise (or [SHIFT] and (TAB] *°r
counterclockwise) through the on-screen scroll boxes and buttons in
sequence until the (CANCEL] button is selected, and then pressing the
[ENTER] key.
Any of these operations will take the user out of the screen in which the user is
working and return to the last active screen before arriving at this screen.
The GloED Main Menu
Scenario Database Calculate Map Reports Help Quit
Figure 1. The GloED Main Menu
The GloED main menu is shown in Figure 1. It is represented by a bar that will
remain at the top of the screen as long as GloED is running, and offers choices of the
type of function the user would like GloED to perform. The user can move between
menu options with the mouse or with the arrow keys. The user selects an item by
clicking on it or by typing the first letter of its name. The menu options and their
general functions are;
Scenario-This menu allows the user to load a previously created
scenario, generate a new scenario, combine elements of two or more
scenarios, or delete a previously created scenario.
Database-This menu allows the user to edit data entered by the user
or to Import data sets to be combined with the system database files.
• Calculate-Commands the software to create an emissions inventory
based on the currently defined scenario.
1-69
-------
Map—Allows the user to display the emissions inventory in the form of a
gridded or thematic map. (Only thematic maps are currently
available.)
Reports-Allows the user to review the emissions inventory in a point-
by-point fashion and to search for a specific data point in the inventory.
It reports the results of the inventory calculation in text form (as
tables) or as graphics (bar charts or pie charts).
Help-Allows the user to receive on-screen assistance while operating
the software.
Quit-Allows the user to leave the program.
The Scenario Menu Option
Scenario^ Database Calculate Mao Renorts Helo Quit I
Generate
rnrnhln*
Delete
Figure 2. Pull-Down Menu for Scenario Options
The user arrives at the step described in this section by choosing the scenario
option on the GloED main menu. The menu that will pull down when the user selects
the scenario option is shown in Figure 2. It offers the option to load a scenario from
the user's disk, generate a new scenario, combine two or more existing scenarios, or
delete a scenario from a disk. By pressing [ESC], or the left-arrow key, the user
closes this pull-down menu and returns to the main menu.
The Scenario Load Screen
After selecting scenario in the main menu, the user can choose the load
option, which will cause a scroll box to come up on the screen. The scroll box will list
all of the scenarios that have been generated in previous sessions. To create a
different report from a previously generated scenario, the user can mouse-click on the
name of that scenario, or use the cursor keys to select the scenario and then press
the [ENTER] key. Now the user can click on the [OK] button and GloED will load the
selected scenario. It is important to remember that the user must now select the
calculate option on the main menu bar before using GloED to generate reports of
this scenario.
The scenario load screen is shown in Figure 3.
1-70
-------
Select an Existing Scenario
Scenario: gEWojiETSCf*^ ¦<;
Scenarios:
NEWONE.SCN
Figure 3. The Scenario Load Screen
The Scenario Generate Screens
If the user selects the generate option in the Scenario menu, GloED will guide
through a series of screens in which the user names, describes, and defines the
parameters of the scenario to be generated. When this option in the scenario menu is
selected, a dialogue box entitled "Create a New Scenario* will appear. In this dialogue
box, the user enters the name of the scenario to be created. The user should enter a
valid DOS file name that is sufficiently descriptive to be remembered if it is to be used
again later. Then [TAB] to or mouse-select on the [OK] button and the Scenario
Description screen will come up to allow entering a description of the scenario to be
generated.
To replace an existing scenario with the one the user plans to generate, press
the [TAB] key or use the mouse to enter the scroll box that appears at the bottom
of the dialogue box. Then use the mouse or cursor keys to select the scenario to be
replaced. When the appropriate scenario is highlighted, press the [ENTER] key and its
name will appear on the scenario name line at the top of the box. Then mouse-click
on the [OK] button or [TAB] to the [OK] button and press the [ENTER] key. Since
this scenario exists, a warning box will appear on the screen that says "This
Scenario existsl Rebuild?" and gives on-screen buttons that say [OK], which
replaces the existing scenario with your new one, or [CANCEL], which cancels the
scenario generate routine. Select [OK] and the Scenario Description screen (Figure 4)
will come up to allow entering a description of the new scenario to be developed.
The Scenario Description Screen
This screen will first ask the user to enter a long description of the scenario to
be generated.
I OK
Cancel
1-71
-------
r
Scenario Description
Description:
This scenario generates an inventory of global NOx
emissions resulting from fossil-fuel production.
Brief Description:
MffiH I
L
OK
Cancel
Figure 4. The Scenario Description Screen
The Unit Conversion Ufllltv-After entering the long and short descriptions
of the scenario, hit the [TAB] key again. The user will now be in the "Units" field of the
scenario description screen, in this field, enter the target units in which the inventory
is to be reported. The user does not need to know the units used by the individual
data sets because GloED contains a unit conversion utility that will automatically
convert the results of the scenario to the units defined as targets. This utility will be
especially useful in comparing the results of a series of scenarios. If entering the same
units for all of the scenarios, no conversion will be required to make a comparison.
The Scenario Generate Screen
Once the new scenario is described to the user's satisfaction, select [OK] and
GloED will lead to the Scenario Generate screen (Figure 5). This screen has four scroll
boxes, one each for data sets, countries, source categories, and pollutants. Move
among the scroll boxes and buttons on this screen, using the mouse to select a scroll
box item or pressing the [TAB] key to move clockwise, or [SHIFT] and [TAB] to move
counterclockwise. The other scroll boxes will show the source categories, countries,
and pollutants that are defined in the selected data set. When the screen first comes
up, nothing will be selected. After the user has selected the data set(s) to use, the
other boxes will fill with the countries, source categories, and pollutants in that data
set.
1-72
-------
Scenario Generate
Data Sets:
Source Categories:
ALL
CLEAR
GLOBAL VOC INVENTORY
BIOGENIC SOURCES
ALL
CLEAR
AGRICULTURE (coal)
AGRICULTURE (gas)
OK
Countries:
Pollutants:
ALL
CLEAR
ALBANIA
ALGERIA
ALL
CLEAR
NOx
OQ2
Cancel
Figure 5. The Scenario Generate Screen
When defining a new scenario, it is important to be aware of the order in which
the items are added to the scroll boxes. When a data set is selected, the other three
scroll boxes will automatically list all of the countries, source categories, and
pollutants included in that data set. GloED defines the items in the other scroll boxes
on the basis of a hierarchy. The countries listed depend upon the data set(s) chosen.
The source categories listed depend upon the data set(s) chosen and the countries
that have been selected within those data sets. The pollutants listed depend upon the
data set(s), countries, and source categories chosen. Thus, every time the selection
of elements in a scroll box is changed, the contents of the other boxes will change
according to this hierarchy.
The user can select a different data set or add data sets to the scenario.
Data sets can be selected or deselected by clicking them with the mouse or by using
the cursor buttons to highlight the desired data set and then pressing the [ENTER]
key. Once the data set(s) are selected, the user can [TAB] or move the mouse
through the next boxes and select the specific source categories, countries, and
pollutants to be reflected in the emissions inventory. ALL automatically selects all of
the items in each list. To select a large number of items, select ALL and then deselect
the few items not wanted in the scenario. CLEAR automatically deselects everything in
the list, and can be used to clear all selections if an error has been made or to select
a few elements in a scroll box. At least one item in each scroll box must be selected
for a scenario to be generated. If the user has selected a combination of data sets,
countries, source categories, and pollutants that does not reflect an actual
combination in the system databases, an error message will request fewer
restrictions on the scenario. When all of the specific items for the desired scenario
have been defined, tab to the [OK] button on the screen, or click on it with the mouse.
GloED is now prepared to calculate the emissions inventory.
1-73
-------
The Calculate Menu Potion
When the calculate option is selected in the GIoED main menu, the system will
generate an emissions inventory based on the current scenario. This is the inventory
that will be used in ail of the following mapping and reporting menu options. Only one
emissions inventory will be generated for each scenario.
The Mao Menu Option
Scenario Database Calculate Map# Reports Help Quit
Gridded >>
Printer
PCX Fife
Figure 6. Options in the Map Menu
Once a scenario is defined and an emissions inventory is calculated for it, the
user is ready to create a map that reflects the results of the inventory. When the
user selects the map option in the main menu, the user wili be given the option to tell
GIoED to display the results either on a gridded map or on a thematic map (see
Figure 6). Gridded maps display the distribution of emissions in the form of ranges on
a latitude/longitude grid. (Gridded maps are not yet available.) Thematic maps
display ranges of emissions on a country basis. Figure 7 is an example of a thematic
map.
When the type of map to be created is selected, another menu wili pull down
and request definitions of the output location for the map. If the user selects Screen,
the map will appear on the computer's video monitor. To send the map to the
printer, a dialogue box will appear on the screen and ask for selection of the available
printer output port to which the file is to be sent and for definition of the type of the
user's printer (either by selecting the name with the mouse or by pressing IENTERJ on
the appropriate output port and printer name). Finally, to save the map as a graphic
.PCX file, a dialogue box appears and requests entry of a valid DOS file name (ending
with .PCX) for the map file. At the end of each of these processes, select [OK] and
the Text Report Priorities screen will appear.
When the user has the inventory appropriately constructed and has selected
the map type, GIoED will prepare the map and then give the option of displaying the
map on the screen, sending it to the printer/plotter, or saving it on a disk.
1-74
-------
Methane Emissions From Rice Cultivation
kg/year
-j
Ul
< 3.000,000
UJ< 8.000,000
< 20.000.000
V
\
< 40.000.000
< 60.000,000
< 200.000.000
< 400.000.000
< 900.000.000
< 30.000.000.000
F re 7. Thematic Map
-------
The Reports Menu Option
Rf^nnrifi Calculate Map tRepor£g2] Help Quit
Pie >
Bar >»
Text >
Export >
Figure 8. The Reports Menu Option
When the reports option in the main menu is chosen, the menu that pulls down
(see Figure 8) gives the following options: Review, Pie, Bar, Text, and Export The
Pie and Bar functions create graphical repohs of the emissions inventory. The Text
function prepares a complete, tabular report of the emissions inventory, and the
Review option allows viewing complete information on individual data points in the
emissions inventory.
The Export menu option is used to save the tabular results of the emissions
inventory. When the user selects this option in the Reports pull-down menu, another
menu pulls down and gives the option of saving the report in Lotus 1-2-3 format,
ASCII format, or dBase III format. Figure 9 is a Reports Export screen for exporting
a GloED data file to dBase.
Save an Inventory to a dBase File
Filename: C:\GLOED\\DBF
Directories Files
-
OECD87.SCM
R1CE.SCM
— A —
... b —
— C —
— D —
ACTIVnY.DBF
ALLCNTRY. DBF
ALLDSET.DBF
ALLPOL.DBF
AIXSCAT.DBF
CATEG OKY\ DBF
COUNTKY.DBF
DATASET.DBF
DSETBUP.DBF
EFACTORDBF
OK
Cancel
Figure 9. The Reports Export Screen
1-76
-------
Graphical Reports
Scenario Database Calculate Mao Reports : | Help
Quit
Review
BfireesHH
Bar >
Printer
Text >
PCX File
Export >
Figure .10. The Graphical Reports Option
In choosing to report the results of the scenario graphically, the user has the
option of presenting them in bar chart or pie chart format. Select Bar or Pie in the
Reports menu and then choose to have the jgraphical report appear on screen, be
sent to a printer, or saved on disk as a .PCX file. In selecting the printer option, a new
screen appears (Figure 11), asking for the printer type and output port for the fife.
When the print location is designated, select [OK] and the title screen will appear. In
this screen, type the primary title for the graphical report, press [TAB] or mouse-click
on the second title field, and enter the secondary title (usually a reference for the
data, the units for the graph, or some sort of explanatory note) for the graph. Now,
select [OK] and GloED will send the report to the printer. Figures 12 and 13 are
examples of a bar and a pie chart, respectively.
Save a Chart to a PCX File
Filename: CAGLOEDV.PCX
Directories Files
OECD87.SCM
RICE.SCM
— A —
— B —
— C
... d —
OK
Cancel
Figure 11. The Reports Menu for Saving a Pie Graph to a .PCX File
1-77
-------
Methane Emissions From Rice Cultivation
Percent Emissions
40
m
c
o
T?
in
U
"3
x>
JD
o
c
V
O
u
V
a.
30
20
10
U
555
Jlllllll.
INDIA
CHINA
THAILAND
BANGLADESH
BURMA
If VIETNAM
^ BRAZIL
j= PHILIPPINES
III JAPAN
OTHER
Figure 12. Bar Chart
-------
Methane Emissions From Landfills
H UNITED KINGDOM
SS GERMANY
= ITALY
III JAPAN
OTHER
I
USA
USSR
CHINA
INDIA
CANADA
Figur 3. Pie Chart
-------
Text Reports
Scenario Database Calculate Map jj&port*A Help Quit
Review
Pie
>
Bar
>
Export
>
Screes
Printer
File
Figure 14. Options in the Reports Menu
To present the results in the form of a table, referred to here as a text report,
mouse-click on the Reports option in the GloED main menu, or type an R to open the
menu. Then select the Generate option in the menu shown in Figure 14 that appears
at the right of the pull-down menu. A dialogue box, called the Text Report
Priorities Screen, will appear. It will allow assignment of totalling priorities to the
pollutants, source categories, and countries that will appear in the tabular report
When the report has been prepared, use the GloED main menu bar to display
the report on the screen, to send it to the printer, or to save it on a disk. To see the
report without printing it, sending it to the screen allows using the GloED Browser
Screen to scan the report, using it like hard copy. The browser screen, showing a
section of a report, is shown in Figure 15. This screen functions very much like the
smaller scroll boxes in GloED except that it has both horizontal and vertical scroll
bars. Mouse-clicking on the arrows at the ends of the bars moves one line per click in
the direction indicated by the arrow. For larger jumps in the report, slide the narrow
bars in the middle of the scroll bars in the desired direction.
1-80
-------
|NOx - France, Canada, U.S. - Fossil Fuels
COUNTRY
CANADA
CANADA
CANADA
CANADA
CANADA
CANADA
CANADA
CANADA
CANADA
CANADA
CANADA
FRANCE
FRANCE
FRANCE
FRANCE
FRANCE
FRANCE
FRANCE
FRANCE
FRANCE
FRANCE
FRANCE
UNITED STATES
UNITED STATES
UNrTED STATES
UNITED STATES
UNfTED STATES
UNITED STATES
UNITED STATES
UNITED STATES
UNITED STATES
OF AMERICA
OF AMERICA
OF AMERICA
OF AMERICA
OF AMERICA
OF AMERICA
OF AMERICA
OF AMERICA
OF AMERICA
SOURCE CATEGORY
AIR (oil)
AUTO OF ELEC (gas)
AUTO OF ELEC (oil)
PUB SERV ELEC (coal)
PUB. SERV ELEC (gas)
PUB SERV ELEC (oil)
RESIDENTIAL (gas)
RESIDENTIAL (oil)
ROAD (oil)
TOTAL INDUSTRY (gas)
TOTAL INDUSTRY (oil)
AIR (oil)
AUTO OF ELEC (gas)
AUTO OF ELEC (oil)
PUB SERV ELEC (coal)
PUB SERV ELEC (gas)
PUB SERV ELEC (oil)
RESIDENTIAL (gas)
RESIDENTIAL (oil)
ROAD (oil)
TOTAL INDUSTRY (gas)
TOTAL INDUSTRY (oil)
AIR (Oil)
PUB SERV ELEC (coal)
PUB SERV ELEC (gas)
PUB SERV ELEC (oil)
RESIDENTIAL (gas)
RESIDENTIAL (oil)
ROAD (oil)
TOTAL INDUSTRY (gas)
TOTAL INDUSTRY (oil)
EMISSIONS
5.406285E+04
1.747200E+01
8.216900E+02
5.616418E+05
5.457480E+03
1.867441E407
1.860558E+04
1.165500E+04
5.729220E+05
4.867460E+05
8.117909E+04
4.206498E+04
7.956000E+03
3.985520E4O3
7.106130E+04
2.044000E+01
8.545440E+06
1.324134E+04
4.969440E+03
5.848920E+05
2.864420E+05
1.024266E-H05
9.154994E405
1.050033E+07
6.912297E+05
2.716832E+08
1.908788E+05
8.908956E+04
6.855030E+06
3.270462E+06
6.462365E+05
3.249974E+08
3.249974E408
Figure 15. The Text Report Browser Screen
To see a specific data point in the scenario, or to review the set of elements
chosen for the loaded scenario, use the Review option in the Reports menu. When
Review is selected in the Reports pull-down menu, the Review Emissions Inventory
screen (Figure 16) appears on the screen. When it first comes up, the screen shows
the emissions of the first pollutant chosen for the first country and source category in
the first database used in the currently loaded scenario.
4.
1-81
-------
Review Emissions Inventory
Dataset
Country:
Source Category:
Pollutant:
Emissions:
Units:
Up
Down
Filter
Review
OK
Figure 16. The Inventory Reports Review Screen
From this point, the on-screen [UP] and [DOWN] buttons allow movement
through the scenario, one point at a time, in the indicated direction.
The [REVIEW} button opens a second window, in which appears the database
from which the data have come, the reference for the data, and the emission and
activity factors from which the emission estimate has been calculated.
If the user already knows the data point sought, or to narrow the search,
select the [FILTER] button, which activates the Search Inventory screen. The Search
Inventory Screen looks and works exactly like the Scenario Generate Screen, but the
scroll boxes on this screen contain only the database(s), countries, source categories,
and pollutants chosen for the currently loaded scenario. As a result, this screen
serves as a reminder of the complete contents of the loaded scenario and as a tool
for performing a very specific search of the emission inventory. To search for an Item
in the loaded scenario, use the [TAB] key or the mouse to move through the scroll
boxes on the screen and select, with the [ENTER] key or the mouse, the parameters
of the data point When the user selects the [OK] button, QloED returns to the Review
Emissions Inventory Screen, now containing the information about the data point of
interest.
RELATED ACTIVITIES AND FUTURE PLANS
So far, EPA efforts have been directed toward development of the GloEO
software arid development of emission factors and activity data for anthropogenic
sources of methane and nitrous oxide. With the completion of GloED software,
population of the software will begin with available information on greenhouse gas
emissions data on a country and source specific basis. This data development
1-82
-------
activity can be thought of as filling a 3-D matrix. This 3-D matrix can be envisioned as
a matrix cube with countries along the vertical axis, greenhouse gases along the
horizontal axis, and sources or sectors filling the third dimension. EPA will begin .
loading the software with information on methane emissions and will complete the
inventory data entry with available information from other sources. After published
data are loaded and quality checked, additional data estimates will enable a global
inventory of all greenhouse gases. It is recognized that some of the estimates will be
based on very weak information. However, data quality will be identified throughout
the matrix. This will be a beginning point for identifying where data quality needs to
be fortified.
Though the GloED emissions database will continue to be the primary emphasis
of database software development throughout 1993, we hope to begin development
of a companion database that will contain information on greenhouse gas mitigation
technologies. This database will be called GloTech. It will be an electronic file cabinet
that will house greenhouse gas mitigation technology and report parameters such as
emissions reduction capability, cost, and date of availability of the technology. When
populated with data, GloTech will allow scenario development and file interaction
similar to GloED. This will enable the user to perform cost effectiveness calculations
for an array of technologies that will be constructed in a scenario. Once the scenario
is constructed the user could then determine total cost, total emissions reduction
performance, and other parameters such as secondary impacts (water, solid waste,
etc.), estimated dates of availability of the technology, and limits to market
penetration. Like GloED, GloTech will allow construction of the scenarios based on
information resident within the software. It will also allow the user to input new data
or to modify data within the databases. GloTech will also have each piece of
information linked to its reference to ensure a clear data pedigree.
After GloED and GloTech are operational and fully tested, they could be linked
so that the user can perform "what if" scenarios. These scenarios would provide
estimates of the effect of implementation of certain technologies on country specific
emissions, and what the cost of those technologies would be.
SUMMARY
GloED is a powerful emissions database handling software package that is
nearing completion. It has been presented to OECD for consideration as a standard
tool for global greenhouse gas emission databases. The GloED software will continue
to be refined and updated, and enhancements are planned to enable GloED to accept
gridded information and to interface with geographical information systems. Parallel
to the development of the software, data population activities will produce a global
emissions database which may help to establish baseline emissions for an international
greenhouse gas emissions treaty such as was proposed at the U.N. Earth Summit in
Rio de Janeiro last June. GloED could also be the mechanism to track progress under
such a treaty by allowing annual updates of emissions information on a country and
sector specific basis.
1-63
-------
The development of GloTech will allow estimation of the costs associated with
greenhouse gas emissions reduction. Finally, the GloED and GloTech software working
in concert will allow the user to construct reduction scenarios, estimate impacts, and
view the results of the implemented technologies.
Reference
1. Intergovernmental Panel on Climate Change. "Workshop on National Inventories
of Greenhouse Gas Emissions and Sinks. 5-6 December, 1991. Geneva,
Switzerland. Proceedings." World Meteorological Organization/United Nations
Environment Programme. Edited by P. Schwengels.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency s peer and
administrative review policies and approved for presentation and publication.
1-84
-------
SESSION II: INTERNATIONAL ACTIVITIES
Jane Leggett, Chairperson
Beyond Rio,
a Dutch perspective on the future of research and policies on climate change
by Hans van Zijst, LL.M.
Counselor for Health and Environment at the Royal Netherlands Embassy in Washington
Mr. Chairman, ladies and gentlemen,
I am honoured to have the opportunity today to give you the Dutch perspective on the future of
research and policies on climate change. A future that started with the worldcommunity signing
the Convention on Global Climate Change in Rio this summer.
The Netherlands has been widely reckognized as being proactive in the field of the greenhouse
gas effect. But -quite frankly- we believe that we have no choice. We are the only industrialized
country in the world that lies for more than 50% below sealevel. We are therefor vulnerable to
climatological changes, specifically a rise of the sealevel. In the National Environmental Policy
Plan of 1989 the Dutch government laid out its plans for the future with a long-term approach to
combat pollution and to implement the Brundtland-concept of sustainable development.
"The Plan introduces the five-scale model to classify the size of environmental problems and to
indicate the geographical size of the solutions. If a problem is global, like for instance the climate
change issue, the solution is global and the level of negotiations and implementation is global.
Therefor the Dutch decisionmakers in the field of global climate change became diplomats.
And they will probably stay that way for the coming years, because having a convention (how
hard the battle to achieve it may have been) is just the beginning of a long process of national
planning, monitoring, research, cost/benefit analyses and maybe renegotiations. It is therefor that
I welcome the opportunity to share our thoughts on the future with you today, i.e. to give prompt
start to the necessary implementation of the Rio result.
Current Dutch policy goals (table I)
The Dutch government has taken measures to reach a stabilization of C02-emissions in
1994/1995 at the level of 1989/1990 (i.e. 182 million tonnes per year). It is intended to achieve an
absolute reduction of between 3 and 5% (i.e. to a level of 173-177 million tonnes) by the year
2000. The long-term goal is reaching the sustainable level according to today's scientific know-
2-1
-------
how in the year 2010 and that means an absolute 20-30% reduction of C02-emissions by the year
2010 as compared to 1989/1990. For this long-term goal no policies have been implemented yet. I
will focus on the measures we are taking for the short-term goal of stabilizing the C02 emissions
in 1995.
We have been focusing heavily on C02 because this greenhouse gas accounts for half the problem
and is relatively well-known in its sources and effects. Additional policygoals for other
greenhouse gases have been developed too. For methane (accounting for approximately 20% of
the greenhouse gas emissions) the policy goal is a 10% reduction in 2000 as compared to the 1990
level. For Nitrous oxide (N20) the goal is stabilization of the 1990 level in 2000. For ozone-
triggers like reactive odd-nitrogen and volatile organic gases the goals are considerably higher,
ranging between 50 and 60% reduction in 2000 as compared with the 1988 level. The policygoal
for CFCs is total banning of all production by the year 1993 and the rapid phasing out of all use
as soon as possible.
Current Dutch policy measures
The current Dutch policy measures are laid down in the Memorandum on Climate Change, which
is generaly accepted as the National Plan under the Convention on Gimate Change. Additional
measures are taken on basis of the Memorandum on Energy Conservation and the
Structurescheme for Traffic and Transport. In choosing from a wide set of opportunities the
Dutch government favors measures that are more cost-effective and those that will have positive
side-effects in other fields. For instance, measures aimed at energy saving, more efficient use ot
energy resources, waste prevention, the reduction of growth in mobility and afforestation will
help to combat acidification and the smog problem as well as promote the better management of
resources and enhance sustainable development.
Before I show you the measures we are taking, please take a look at table 2 on sources of C02
emissions in the Netherlands. As you can see Industry and Power plants take care of almost half
of the emissions. Another interesting feature is that only households have succeeded in reducing
their emissions between 1985 and 1988. The fastest grower is waste incineration. Although only
accounting for 1% of the emission, it increased with almost 25% over the three year timeframe.
2-2
-------
The measures we are currently taking are shown in table 3. The measures are divided in four
categories:
* Energy saving
* Fuel consumption
* Traffic and transport
* Waste sector.
In the energyfield we must achieve an average improvement of 2% energy efficiency per year
over the last decade of the 20th century. It can be achieved by insulation of existing buildings
and houses as well as sharper energy standards for new buildings and by increasing the energy-
efficiency of Dutch industry. Instruments are among others subsidies on investment in new
technology and on high-efficiency central heatingsystems, information transfer and consumer
awareness building.
In the fuel consumption field the decreased use of coal for electric utilities and industries are
heavily advocated. In part coal can be replaced by natural gas, but also solar and wind energy
have a high feasability for a considerable group of industry. Investment subsidies and
demonstrationprojects are the most important instruments.
In traffic and transport the biggest help comes from the almost gridlocked situation on the Dutch
highway system during rush hours. By improving public transport (which compared to the US is
already superb) and enforcing growth limitation to car traffic the goals for traffic and transport
could be achieved. In addition transportmanagement by companies and the improvement of
driving behaviour (speed as well as car maintenance) can contribute to achieving the goals. Part
of the solution must come from European legislation on fuel efficiency standards for cars. Fiscal
measures to favor smaller and cleaner cars are considered.
In the waste sector policies are put in place to considerably lower the amount of waste that has to
be incinerated. Especially compost forming and anaerobic fermentation make that possible.
Furthermore we are thinking about commercially extracting methane gas from waste dumps.
Not on this table but also an integral part of the Dutch approach is afforestation. In accordance
with the outcome of the 1989 Ministerial conference in Noordwijk, The Netherlands has pledged
to plant about 3500 hectares per year. In addition to that the Dutch Association of Electricity
Utilities has taken the initiative to reforest in Eastern Europe and in tropical areas (which is more
cost-effective than taking domestic measures).
2-3
-------
C02 and/or energytax
On top of these measures the Dutch government is a strong advocate of the OECD-wide
introduction of an energytax on fossil fuels. The Dutch proposals and the report of the Dutch
Wolfson Commision have been negotiated dramatically inside the Netherlands. The economic
effects of unilaterally introducing such a fiscal measure can be strong, even if macro-economic
calculations show that recycling the assets in the economy for environmental and energy-efficient
purposes can have a positive effect on the Gross National Product. On a micro-economic level
distortions can take place, if the tax is equajly applied to all industries and utilities.
For the time being we wait for the negotiations in the coming months of the proposal of the
European Commission.
Is it enough?
Although the Dutch approach is in our own eyes rather comprehensive and for the short term
hopefully will show to be sufficient, it will not be enough. Greenhouse gas emissions are a long
way from the sustainable level. According to the first evaluation of the Netherlands
Environmental Policy Plan, called 'National Outlook 2, 1990-2010* the Dutch Research Institute
for Public Health and the Environment concludes that further dramatic reductions will be
necessary in the first decade of the next century. The forecast shows a setback in our current
expectations for 2010. To justify stronger measures science will have to move on in clarifying the
process of global warming, indicating the sources, the interaction of human and natural sources,
explaining the role of sinks like trees and oceans, explaining the role of clouds and other
climatological phenomena. And politicians will be asked to decide on basis of the outcome of that
research, even if the answer is that science doesn't know yet.
Therefor we strongly support the continuation of the IPCC work, the International Geosphere
Biosphere Programme, and similar international efforts as well as we continue and fortify our
national research program on climate change. Our national efforts shall shift its focus to the
development of policy modelling, long-term qualitystandards and scenario's for sustainable
2-4
-------
solutions (with technical, economical, socio-psychological, administrative aod regulatory options).
We will also support developments in the field of monitoring, like the remote sensing program of
the European Space Agency.
International cooperation
Apart from the evident need to share international knowledge of science there is the need of
international cooperation in implementing the Convention on Climate Change. As I said earlier
Rio is only the beginning and we must push for a prompt start to keep the momentum going or
create new momentums in the near future, otherwise the chemistry of the UNCED-summit will
be gone. The call for national plans must be transformed in the drafting of these plans. The
Dutch government will strongly support the call for speedy ratification and implementation of the
convention in domestic policies, in the developing as well as in the OECD-countries. We will
support the reconvening of the INC-meetings for further clarifying of some open points in the
convention and to speed up the implementation process. The work of the IPCC must be continued
and among other things could be focused on the development of internationally agreed
methodologies and reporting formats for emission inventories. In that framework I am happy to
say that the Dutch Research Institute for Public Health and the Environment will host an I FCC-
workshop in early February next year in Amersfoort, The Netherlands. The workshop is about
Methane and nitrous oxide and will focus on methods in national inventories and options for
control. Those of you who want to have a copy of the first announcement can give me your
business card and I will make sure that you get a copy.
4.
2-5
-------
Table 1
Policygoals for GHG's
C02
Baselevel 1989/1990 182 MT/Yr
Stabilization 1994/1995 182 MT/Yr
Reduction 3-5% 2000 173-177 MT/Yr
Sustainable level 2010 127-145 MT/Yr
CH4
Baselevel 1990
Stabilization 1995
Reduction 10% 2000
960 KT/Yr
940 KT/Yr
840 KT/Yr
N20
Baselevel 1990
Stabilization 2000
40 KT/Yr
40 KT/Yr
Ozone-related gases (CO, NOx, VOG)
Baselevel CO 1990
Reduction 2000
Baselevel NOx 1988
Reduction 2000
Baselevel VOG 1988
Reduction 2000
1100 KT/Yr
540 KT/Yr
560 KT/Yr
240 KT/Yr
490 KT/Yr
205 KT/Yr
2-6
-------
Table 2
Sources of C02 amissions in the Netherlands
(1988 figures)
Source
Share of total
Trend 1985-1988
Industry
24%
+
0.7%
annual
Power plants
22%
+
4.5%
annual
Traffic
16%
+
2.5%
annual
Private hones
14%
-
1.5%
annual
Refineries
8%
+
10.4%
annual
Warehouses
5%
+
13.4%
annual
Waste incineration
1%
+
23.3%
annual
Other sources
10%
0.7%
annual
Total
100%
+ 3.3% annual
-------
Table 3
Concrete measures in implementation to decrease level of C02
emissions
Energy saving * Improved insulation of existing homes
and other buildings
* Increasing the energy efficiency of
machines
* Investment in energy saving
* Construction of homes and other
buildings with low energy consumption
Fuel consumption * Decreased coal use in power plant
sector
* Disincentives to coal use in industry
Traffic and transport * Improvements in public transport
* Car traffic growth limitation
* Transport management in companies
* Improvements in driving behaviour
Waste sector * Compost forming
* Anaerobic fermentation
* Recycling of energy-intensive products
* Methane gas extraction from waste dumps
* Gasification of synthetic materials
The work described in this paper was not funded by the U.S. Environmental Protection Agency. The
contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred.
2-8
-------
SESSION III: C02, EMISSIONS, CONTROL, DISPOSAL AND UTILIZATION
Ken Freidman, Chairperson
Carbon Dioxide
Sequestration
by
Robert P. Hangebrauck. Robert H.
Borgwardt, and Christopher D. Geron
Air and Energy Engineering Research Laboratory
ABSTRACT
Mitigation of global climate change will require the stabilization
of atmospneric concentration of greenhouse gases, especially
carbon dioxide (CO,). CO; can be sequestered by flue-gas and
fuel CO; sequestration or by atmospheric C02 fixation/utiliza-
tion. Flue-gas sequestration involves separation/concentration,
transport, and either disposal or use. Disposal options are either
land or ocean based. Utilization is by either chemical or
biological utilization (recycling). Flue-gas-ariented techniques
in general have high economic and energy costs, but a few areas
show potential ana warrant research ana development (R&D)
attention, especially those holding promise of combined CO/
sulfur dioxide (SO^nitrogen oxides (NOj) control and the inte-
grated gasification combined cycle approaches. CO, disposal
is neither a "sure thing" nor a permanent solution, with options
needing further environmental assessment. Near term, some
CO, recycling is possible, and R&D to examine longer-term
prospects seems warranted. Atmospheric CO, fixation/utiliza-
tion involves either enhanced terrestrial or marine fixation with
utilization of the biomass in some cases. Atmospheric fixation
approaches which seem most attractive are those involving
enhanced biomass C03 sequestration combined with utilization
of the biomass for energy to displace fossil tuel. Of these the
most attractive for R&D appear to be advanced direct combus-
tion using biomass and use of biomass as a source of hydrogen
to leverage fossil fuel use for methanol production (Hydrocarb
process).
Introduction
Internationa] scientists working on the problem of
global climate have concluded that increasing concen-
o 290
2000
Figure 1. Atmospheric CO} increase in the past 250 years
(IPCC, 1990). (Reproduced with permission.)
Paper 3-B
trations of greenhouse gases, especially C02 (Figure 1),
are leading to global warming (Figure 2). The pre-
dicted extent of such warming is the subject of substan-
tial research with model-dependent estimates having
considerable uncertainty . The Intergovernmental
Panel on Climate Change (IPCC) 1990 estimate was
0.3 0 C/decade with a range of 0.2 to 0.5 ° C/decade. The
1992 IPCC Supplement implies a warming rate at the
lower end of this range. Warming has also been studied
2 -0 *
1870
1S90
1970
1990
1910 1930 195
^ YEAR
Figure 2. Global-mean combined land-air and sea-
surface temperatures, 1861 -1989, relative to the
average for 1951 -80 (IPCC, 1990). (Reproduced
with permission.)
based on past temperature records, and has been
determined to be on the order of 0.3 to 0.6 0 C/decade
since the turn of the century. This trend is illustrated
in Figure 2. The abundance of fossil fuels and their
relatively low cost will probably ensure their use as a
principal energy source for the foreseeable future.
Once decisions are put in place to utilize fossil fuels in
the energy infrastructure, long-term commitments will
have been made to release C02 emissions to the atmo-
sphere. In the absence of constraints, the CO, emission
rate can be expected to continue the rapid growth
indicated in Figure 3. This suggests the need for
6000 ¦
5000 •
c
©
A
k
«
U
O 4000-
3000 ¦
e
o
'2
4-ft
4>
E
c
o
2 looo-
2000 ¦
j
\f
f
i
/
/
i
a r
V
~ *•
1860 1880 1900 1920 1940 1960 1980 2000
Year
Figure 3. Annual global CO; emissions for fossil fuel
combustion, cement production, and gas flaring, 1860 to
1988 (ORNL, 1990).
i. EPA August, 1992
3-1
-------
prudent and timely R&D to provide a technological
base for preventing the increasing global emissions
consistent with economic growth.
Evaluating Potential and R&D Needs of
Sequestration Options
The genera] options for C02 sequestration, summa-
rized in Table 1, include both flue-gas C02 sequestra-
tion and atmospheric C02 fixation in the biosphere.
Flue-gas C02 sequestration seeks to remove and dis-
pose of carbon as a part of fossil fuel use and requires
several steps to be undertaken: C02 must first be
separated and/or concentrated to obtain an economi-
cally handleable volume, which is then transported,
either to a point of disposal or to a point of utilization.
Disposal may be on the land or in the ocean. Utilization
can involve direct use of C02 as a product (e.g., en-
hanced oil recovery) or it can be reacted to a product
such as a hydrocarbon or alcohol. Atmospheric COs
fixation/utilization on the other hand is aimed at en-
hancing or accelerating natural C02 fixation processes
via either terrestrial or marine routes. These include
forest management and ocean fertilization, followed by
utilization for energy or other products.
Table 1. General options for carbon dioxide seques-
tration.
* Flue-gas & fuel C02 sequestration
- Separation/concentration
- Transport
- Disposal
- Land or sea
- Chemical/biological utilization
(recycling)
• Atmospheric C02 fixation/utilization
- Terrestrial
- Marine
A variety of technology assessment factors are signifi-
cant for evaluation of the options for C02 sequestration
(Table 2). The status of a technology can be measured
by its level of development relative to commercial
application. Mitigation potential (possible carbon se-
questration) for a particular technology can be charac-
terized in part by measures such as breadth of appli-
cability and applicability to developing countries. Ap-
plicability to developing countries is an estimate of the
potential for use where energy infrastructures are
evolving and economic and other factors differ substan-
tially from those of the industrialized world. The cost-
effectiveness assessment factor can be measured by an
engineering estimate of the cost per metric ton of C02
emissions prevented.
Table 2. Technology assessment factors for C02
sequestration options.
¦Status of the technology
- Level of development
• Mitigation potential
- Breadth of applicability
- Applicability to developing countries
• Cost-effectiveness
- $/metnc ton of COs emissions prevented
* Environmental and energy considerations
- Potential for adverse effects
- Product versus disposal
- Natural resource use
- Earth surface area required (land use)
- Additional fossil energy required
* Probability of success and R&D requirements
Environmental and energy considerations can be
gauged by the potential for adverse environmental
effects, whether or not an approach results in a useful
product or waste requiring disposal, natural resource
use, and land use. Approaches requiring large surface
areas of the Earth for implementation need examina-
tion for social and ecological impact. The additional
fossil energy requirement for implementation needs to
be examined as a rough measure of overall life cycle
efficiency. Finally, the probability of success and R&D
requirements need to be assessed.
Flue-Gas and Fuel COt Sequestration
A number of approaches are possible for separation/
concentration of C02 from fossil-fuel combustion
sources. Figure 4 illustrates some of the possibilities
including various flue-gas scrubbing approaches, oxy-
gen (02)/C02 combustion with C02 recycle, and coal
gasification combined cycles with carbon monoxide
(CO) shifted to hydrogen (H2) and C02.
The categories of flue gas scrubbing (C02 separation/
recovery) include absorption processes, adsorption pro-
cesses, cryogenic separation, and membrane separa-
tion. Of the flue-gas scrubbing approaches, amine
scrubbing is already in commercial use on a relatively
small scale, but has been demonstrated on coal-fired
power plants as large as 800 metric tons per day (TPD)
(Barchas, 1992). Current application is for enhanced
oil recovery (EOR) and production of soda ash, urea,
3-2
EPA August, 1992
-------
ocean injection. It has potential for
NO, and S02 removal. Re-release
of C02 and other environmental
problems associated with utiliza-
tion/disposal are not quantified.
Mitigation potential includes the
utility sector (new and retrofit).
Small-scale pilot and demonstra-
tion work is underway with Japan
planning a demonstration. In
Canada Saskatchewan Power is
planning a 150 to 350 MW demon-
stration. Canada is also research-
ing land disposal of C02: 1) in de-
pleted hydrocarbon reservoirs start-
ing at 50,000 metric TPD, and 2)
EOR utilization at 6000 to 8000
metric TPD (Sypher: Mueller Inter-
national, 1991). CanadianEOR C02
use will be expanded substantially
in 1992. The likelihood of success is
and methanol with Kerr McGee and Fluor Daniel as high, but applicability depends on the ability to mini-
major suppliers of the technology at a cost of $40/ton of mize air in-leakage and environmental acceptability of
C02 recovered (not including transport, disposal, or C02 disposal. An engineering analysis of a 500-MW „
power replacement). Japanese research (Kansai Elec- powerplantwasdiscussedbyBritishCoalattherecent
trie Power) is focused on this approach which they First International Conference on Carbon Dioxide Re-
think can operate with only 10.8% of boiler energy moval in Amsterdam (Cross, 1992). Another analysis
output for COj recovery and liquefaction (Suda, 1992). was presented by Air Products (UK) (Allam, 1992).
Potential exists for. overcoming some of the efficiency These studies indicate that 02-corabustion/C02 recycle
losses through heat recovery (Steinberg, 1991), but in will be 20% cheaper than amine (MEA) systems, can be
general, investigators report discouraging cost and retrofitted to existing power plants, and would result in
energy prospects. Smelser (1991) estimates that a lower power efficiency loss (8.6%). Capital cost for a
grass-roots pulverized coal plant with 90% CO, re- new plant would also be lower than for a conventional
moval can do much better than a retrofit; however, the PC plant. Cost was estimated at $49 per metric ton of
heat rates are 12.7 versus 15.7 MJ/kWh (12000 versus C02 removed. Distillation of liquefied C02 to recover
14900 Btu/kWh). The flue-gas scrubbing approaches, S02 could produce credits that would reduce cost to
of all categories, such as sorption, cryogenic separa- $16/metric ton.
tion, and membrane separation, seem to share the
potential problems of high cost and substantial de- Considering that this process or flue-gas scrubbing
creases in power plant operating efficiency, but oppor- processes may have the potential for inherent S02 and
tunities seem to exist for research in such areas as NO, control, it is of interest to compare the costs of CO,
cryogenics and membranes (Smith, 1991). For ex- control with that for S02 plus NOx control on a cost per
ample, cryogenics might offer the potential for com- metric ton of carbon basis. This is done in Figure 5 for
binedS02, NO,, and C02 control. Accepting lower C02 low and high ranges of cost per ton ofSOj.NOj, and C02
removal efficiencies may also reduce cost. removed. Costs for combined flue gas desulfurization
(FGD) and flue gas treatment (FGT) are put on a per
COj/02 combustion with C02 disposal is a medium- metric ton of carbon basis and compared with the cost
term option and appears to need engineering evalua- of C02 control on a per ton of carbon basis. High- and
tion along with substantial pilot-scale development low-range costs are assumed for FGD, FGT, and C02
work (Wolsky, 1991). Moritsuka (1991) concludes that control. However, if S02 and NO, control were not
such systems will be more cost-effective than flue-gas required or already in place, then this comparison
scrubbing approaches. Herzog (1991) concluded that would have little meaning. Even with theoffsets in cost
COj/Oj combustion would require the least incremen- for possible inherent S02 plus NO, control, a substan-
tal energy. This approach requires C02 utilization via tial cost increase for C02 control can be seen. Overall
EOR, or disposal via depleted petroleum formations or plant energy efficiency is reduced for the flue gas
3-3
14* CM
•Amine sou
= I • Membrane separali
Adsorption
COAL-fWEO
POWER PLANT
100% C02
Compnsslon/Hquatactlon
100% C02
oy
' COAl-FIRED
power plant
WITH 02/C02
\COMBUS710N
02 -£=
iMlin*
100% C02
PMui*
Disposal
DISPOSAL IN
DEPLETED
GAS/OIL
RESERVES
INTEGRATED GASIFICATION
COMBINED CYCLE (KjCC)
DISSOLUTION IN
SEAWATERAT
700-1000
METER DEPTH
SEA OEPOSmON
AS CO*LAKESAT
»1000-METER
DEPTH
Figure 4. Flue-gas CO; sequestration options with C02 disposal
in the ocean or aepleted gas/oil reserves
EPA August, 1992
-------
c
I
c
~
I
s
o
300
200
s
- s
» 2
•3 *•
f •
1"
100 --
O
T> O
u O
~ FQO (S02) ~ FGT
(NO«)
¦ FGC02 (win
ttinarant S02/NO*
coratoi)
Low High
Range Range
Figure 5. Comparison of costs lor S02 and NO, with the
costs of CO, control on a per metric ton of carbon basis.
{FGD and FGD cost ranges are derived from Emmel. 1990).
sequestration options compared to conventional boiler
without COj control. A similar analysis for energy use
offsets could be done by comparison to plants using S02
plus NOt control, but is not done in this paper.
Integrated gasification combined cycle (IGCC) with
CO, disposal is another option shown in Figure 4. One
conceptual design is based on modification of conven-
tional gasification combined cycle plants with added
steps to shift all CO to C02 for separation. The result-
ing hydrogen-rich fuel when burned in a turbine suf-
fers efficiency loss. An alternative design would in-
crease efficiency by applying C0J02 combustion to the
turbine combustor. No design studies are available to
determine the potential of the latter approach. Mitiga-
tion potential would include new utility applications
only. Canada is planning to investigate the approach
at pilot scale in conjunction with the design of a planned
250 MW IGCC plant (Sypher:Mueller International,
1991). The feasibility study was completed in 1991,
and the plant was scheduled to be in operation by 1996.
The likelihood of success seems to be reasonable, but
the process needs pilot and demonstration work.
Several innovations of the IGCC concept were reported
at a recent Amsterdam Conference on CO, removal.
They were mainly centered on the development of a
high temperature membrane for separation of CO and
H, from the gasifier. The separated gas streams are fed
to different turbines, one of which is fed with pure 02
and recycled CO,. The combustion gases are then fed
to boilers and steam turbines. The exhaust from the
CO, turbine is collected (less recycle) for disposal. This
unit is expected to operate at 35% efficiency and 88%
CO, reduction with the cost of electricity increased by
30%. Capital cost is $1100/kW; CO, recovery cost is
$16/metric ton (Hendricks, 1992). Replacement of the
turbine of an IGCC plant with a molten carbonate fuel
cell can increase the generating efficiency to nearly
48% while removing 90% of the CO,. The fuel cell is
able to convert CO to CO, using air without mixing N,
with the oxidation products. Consequently, no air
separation plant is needed. A demonstration IGCC
plant is un der construction in Germany (RWE Energie)
and will start operation in 1995 using brown coal.
Efficiency is expected to be 45% without CO, recovery
and 38.6% with recovery of 86% of CO, emissions.
Investment cost without CO, recovery would be 10%
less than for a conventional plant; with recovery, it is
30% more. A systems study by British Coal concluded
that the IGCC system is the best choice for the UK,
especially if operated under pressure so that the CO,
can be absorbed in seawater for disposal without lique-
faction. Cost of pipelines for liquid CO, disposal in deep
oceans was judged to be prohibitive. A thermodynamic
analysis of the IGCC system indicates thatMEA scrub-
bing after combustion is preferable to air separation
prior to combustion. Steam for CO, stripping can be
extracted from the turbine. They conclude that the
system is technically and economically feasible at an
increased power cost of $0,015-0.02 per kWhr.
Fuel CO, (carbon) sequestration has been evaluated
extensively by Steinberg (1991). These are methods
that basically remove carbon from a fuel like coal
resulting in a hydrogen-enriched fuel. The removed
carbon is then sent to storage/disposal. The problems
with carbon disposal can be minimized by using a
variation of this approach, called the Hydrocarb pro-
cess, where biomass is used as a source of H,. This is
discussed later with the atmospheric fixation ap-
proaches.
CO, Disposal
Qgftan disposal Any of the flue-gas CO, sequestration
approaches discussed above, implemented on a mas-
sive scale, require consideration of disposal of CO, in
the ocean as well as on land. Although ocean disposal
has been and is being studied ( Hoffert, 1979; Herzog,
1991; Wilson, 1992), it may only be semi-permanent
storage and has enough environmental issues to make
the option longer term. The ocean options considered
include disposal as liquid, solid, and gaseous CO,. Key
practical considerations are economic means for get-
ting massive amounts of CO, down to depths where it
will stay or sink. Solid CO, or CO, hydrates (clathrates)
will sink on their own. Smith (1991) suggests that the
EPA August, 1992
3-4
-------
2000
Lcno-wnn rcmui
« (troapMne CO
i
2100
3500
iooo
Utr
a100% into the atmosphere
b 50% at 1500 m depth into the ocean
c 50% at 4000 m depth into the ocean
d 100% at 1500 m depth into the ocean
e 100% at 4000 m depth into the ocean
Figure 7. C02 C lath rate lattice has approximately 6
molecules of C02 to 44 ol HaO (Miller, 1974).
(Reproduced with permission.)
need to be resolved include:
Figure 6. Modeled atmospheric CO, resulting from use of all * Potential for re-release of C02 via up-welling of
fossil fuel over time with CO, disposed of at various depths
costs for the solids disposal approach make it a less
likely option. Liquid C02 needs to be injected below
1000 m and -- to reduce re-release -- 3000 to 4000 m is
seen to be safer. Scrubbing of stack gas with sea water
and reinjection in the ocean below 1000 m might be
adequate. However, none of these options may really
be permanent. As shown in Figure 6, some part of the
C02 will return to the atmosphere as a function of
disposal depth and time (Hoffert, 1979). However,
reaction of C02 with calcium carbonate in the ocean
bottom sediments could fix some of this C02 as the
bicarbonate and thus reduce the potential for return to
the atmosphere (Wilson, 1992). Another estimate for
the residence time of C02 deposited at a 1000 m depth
is 200-300 years (Liro, 1992). Modeling of projected,
long-term atmospheric C02 concentrations appears to
be a major research need.
Formation and stability of C02 clathrates are cur-
rently the subject of additional research. Clathrates
can form at various depths depending on temperature.
In arctic regions they can form at depths as shallow as
70 m. Relatively little fundamental data are available.
Its density is somewhat uncertain but appears to be
1.02 to 1.12 times that for sea water; therefore, it sinks.
Large amounts of C02 can be tied up per unit volume of
clathrate. Figure 7 shows the structure of C02 clath-
rate. About 6 molecules of C02 are combined with
about 44 of H20 in the clathrate crystal lattice. It has
appearance of wet snow. Nishikawa (1992) has deter-
mined some additional phase equilibrium data and pH
data for sea water, but finds that long-term testing will
be needed to obtain dissolution information.
Ocean CO, disposal environmental issues which
C02
• Conversion of high density hydrates to low density
• Catastrophic releases in transport or storage.
• Potential that long-term circulation could return
C02 to atmosphere.
• Potential for local/regional acidification of ocean
with biological impacts
• Chances for reduction of biological diversity in the
column and ocean floor.
• Potential for accumulation of clathrates on sea-
floor would impair normal biological processes in
and on the sediment and cause disturbance of
bottom sediment
• Potential for local extinction of animals beneath
the deposited material
A report by Golomb (1989) includes a discussion by
Woods Hole researchers of some potential biological
impacts.
The cost of a pipeline for deep ocean disposal appears
to be a major obstacle to this option. A preliminary cost
estimate by British Coal for a 250 km pipeline for a 500
MW power plant is about $270 million dollars or 30% of
the capital cost of the power plant. British Coal favors
the depleted oil and gas field option although it will also
require pipelines (presumably shorter).
On-land disposal. Options include disposal in depleted
oil/gas formations, salt domes, aquifers, and other
geological voids. Smith (1991) estimates a C02 capac-
ity for depleted gas fields worldwide at 1.3 GtC with
new capacity becoming available at 0.7 GtC/year. On-
land disposal is less costly than ocean disposal, espe-
cially inland. We do not know for sure if it is more
secure than ocean disposal and the extent of re-release.
The potential exists for use in EOR, but economic
EPA August, 1992
3-5
Page 5
-------
feasibility in terms of credits for CO, may be
limited under conditions of low oil prices
Also the capacity seems limited, probably
amounting to only a small fraction of a GtC/
yr. After a few years of injection,CO, is likely
to show up at the well head. We do not know
the extent of re-release of CO, used for EOR or
the potential for prevention of re-release, but
it is conceivable that measures could be taken
to prevent leaks and recycle the CO,. Other
issues arise where CO^/SO/NO, mixtures
would be injected. These include the trans-
port of the COj/SOj/NO, mixture over long
distances, the disposition of the SOj/NO, con-
taminants injected in oil reservoirs, and the
potential for pluggage of the reservoir (Spar-
row, 1988). We do not know the potential for
catastrophic release. Disposal in aquifers is
another option currently being discussed, but
questions on the resulting underground chem-
istry and fate remain to be resolved.
CO, Utilization
:
SatalliiB-bMad
toiar photovoltaic
panala (lutuM)
MAN-MADE CARBON CYCLE
fopen c>de)
sei
COAL-FIRED POWER;
• [PLANT WITH 02/C02'^
- COMBUSTION-^
COAL-»sj '
^ 10O%CO2^ re^^r'
s-3
Figure a. Long-term integrated systems for carbon recycle.
Production
Tank*—
ynthesIsPlan
In the near term, CO, utilization is the most available
and environmentally unquestioned option. A good
example is the emerging practice in the chemical in-
dustry of co-siting CO,-using processes with CO,-pro-
ducers -• specifically, integration of methanol produc-
tion (a CO,-using process) with ammonia production (a
concentrated CO.-production process). What is the
longer-range importance of carbon recycling? Carbon
is a key transport agent for H,; e.g., methane (CH4) and
other hydrocarbons provide a "natural," convenient,
and practical means for H, energy transport and use.
On the other hand, H2 is difficult to store, transport,
and use directly. The world's fossil fuel reserves are the
most economically available, concentrated source of
carbon, but are rapidly being consumed. Carbon re-
sources can be preserved as an energy transport me-
dium for future generations by recycling existing con-
centrated sources of CO,. At the same time, C02
emissions to the atmosphere can be reduced along with
the attendant global wanning. What are some of the
currently wasted CO, resources which could be man-
aged for synthesis of future hydrocarbons? They in-
clude: 1) fossil fuel combustion (e.g., coal-fired power
plants), 2) calcination of limestone, 3) oxygen-blown
blastfurnace gas, 4) natural gas acid gas stripping, and
5) ammonia production toname a few. Carbon (as C02)
can also be recycled from carbonate rock or from the
atmosphere to produce synthetic fuel but at much
greater cost (Steinberg, 1977).
Figure 8 illustrates some longer-term integrated sys-
tems for carbon recycling. Utilizing CO, via a "man-
made" carbon cycle needs overall assessment and cost
feasibility studies. It is a chemical-based recycling
approach with no CO, disposal required. This makes
it environmentally attractive. H, (with 0, as a by-
product) can be produced by using land-based solar-
photovoltaic produced electricity to electrolyze water.
For projected huge energy demands by the year 2100,
satellite- or lunar- based photovoltaic generation may
be required (Hoffert, 1991). CO, is reacted with H, to
form methanol. Methanol can be further reacted via
dehydration to produce gasoline if desired. Steinberg
(1977) investigated a similar concept using nuclear
power as the energy source. The system eliminates
need for an air separator and CO, disposal. Mitigation
potential includes the utility, industrial, and transpor-
tation sectors (new and retrofit). If shown to be fea-
sible, the high mitigation potential and elimination of
the CO, disposal problem argue for R&D to reduce H,
production cost Recent breakthroughs in production
of solar photovoltaic cells should help this somewhat,
but innovative means for producing H, is a fertile
research area. Credits for SOj/NO, control could help
to reduce costs. Feasibility as a system needs to be
determined. The Japanese are pursuing R&D on
system components (Arakawa, 1992a&b). CO^O,
combustion is in the early stages of development(small-
scale pilot, small demonstration) as discussed previ-
ously. The likelihood of success depends on key compo-
nents, including low-cost photovoltaic electricity gen-
eration, H, production, and CO/O, combustion. If this
long-range concept proves to be feasible it could have
several advantages: 1) fossil fuel can continue to be
3-6
EPA August, 1992
-------
used, 2) transportation and gas tur-
bine fuels are made available, 3)
fossil carbon reserves are extended
for future use, 4) the energy storage
problem and global transportabil-
ity problem associated with solar
energy are solved, 5) the approach
appears highly applicable to devel-
oping countries, 6) solves the prob-
lem of difficult storage, transport,
and use of H,, and 7) provides a
transition to a direct-use H2 economy
by building up the necessary H,
production infrastructure. Ulti-
mately, the "COj-synfuel" being pro-
duced can be recycled itself, thus
achieving a closed cycle.
Flue-Gas Sequestration (Biologi-
cal Processes)
Other
Products
COAL-FIRED
POWER
PLANT
Alga*
Pond
\ ]] BIOMASS-FIRED
POWER
PLANT
Light Concentrat
Photobioreactor
(Light-Pipe (optic fibor) Reactor
ng Super Ajgae)
Figure 9. Biologically based flue-gas C02 sequestration processes
(with microalgae flue-gas C02 removal).
Flue-gas sequestration via microalgae flue-gas C02
capture is illustrated in Figure 9. The concept is
attractive, but difficult to implement because of the
large surface area required for exposing microalgae to
C02 in the flue gas. A 100-MW power plant would
require an algae farm surrounding the plant out to a
distance of 4.3 km (Brown, 1990). Direct use of the
biomass generated appears to be difficult, but it can be
processed for lipids or to other hydrocarbon fuels such
as methane. Mitigation potential exists for new and
retrofit utility applications but is probably limited to
only certain areas with enough land, proper terrain,
nutrient capital, and adequate water and evaporation
rates. No reliable estimates of costs are available.
SERI is doing bench scale research and the Japanese
are looking at photobioreactor concepts which would
allow more efficient and concentrated
biological growth.
options in perspective, however, consider, that if the
entire U.S. current annual wood production (net growth
of 493 million dry metric tons/year) were diverted to
wood energy, it optimistically could provide only 7 x
10" J (7 quads). Figure 11 is based on wood production
on U.S. timberlands as projected by Sampson (1991). It
is likely that only a fraction of these amounts would be
used for energy. Approximately 1.5 x 10" J (1.4 quads)
is used for energy currently (High, 1990). Only a
fraction of the land area of a country is suitable, by
climate or terrain, for energy plantations. Within that
fraction, only an area within a reasonable distance of
an energy plant can be considered useful because of the
high transportation costs of biomass (one trainload of
coal is equivalent to the energy content of eight train-
loads ofbiomass). In any even t it appears that the most
Atmospheric CO, Fixation
Enhancing Terrestrial Fixation
Figure 10 illustrates some of the op-
tions for enhancing terrestrial fixa-
tion. Options include reforestation
and forest management (0.1 - 1 Gt/
yr), reducing deforestation (1 Gt/yr),
and enhanced soil C sequestration
(0.1 -0.5 Gt/yr). While there do not
appear to be any "silver bullets" in the
options available, it makes sense to
continue efforts underway to maxi-
mize potential benefits. To put these
ATMOSPHERIC CO., (750 GtC)
ii
•5 o
£2
li
ir"-
Forests
(Global)
I I
Agroforestry &
Sustainable Agri.
(Tropics)
2
at
M
— E
£f
li
Soil Organic C
(U.S. A Temperate)
Reduce rale af
deforestation by
agricultural clearing
Increase ca/bon ilorage
in organic traction
Inaeoi C-pooiby
expanding a/ea.
increasing stocking
(HeloresiatiofVAI(orestation)
•Low promise, land eon flics
Figure 10. Means for increasing atmospheric CO, fixation via enhancement of
terrestrial system uptake. Units are metric gigatons of carbon (GtC).
* EPA August, 1992
3-7
-------
10 J (Quads)
10
lor ftWMn
15
Current
production
Mtvunol '
Methanol
Synthesis
2CO°*C
SOU# P* (SOMn.
Hydro-
asificatlon
sactor
900 "C
SOiM^ Pa (SO «»n.
Pyrolysis
Reactor
Potential future
production
Carbon
Figure 11. Energy from biomass if total wood production
were diverted to energy: 1) all current U.S. wood production
and 2) all potential future U.S. production (not including
dedicated energy plantations).
effective approach for taking advantage of terrestrial
fixation is to use the biomass resources produced to
produce energy to displace fossil energy (Hall, 1992).
Lee (1991) estimates that short rotation energy crops
could provide an additional 6 to 13 x 1018 J (6 to 12
quads). Vitousek (1991) finds that new tree planta-
tions without energy use can at best cause a brief delay
in CO, accumulation. He also points out the concerns
that 1) the biomass truly replaces fossil fuel and not
just simply increases energy available to energy defi-
cient people, and 2) energy plantations be developed on
land that is not now forest covered because of a sub-
stantial net loss of carbon during the transition. Better
yet, approaches like the Hydrocarb process, if success-
ful, may leverage available biomass by using fossil
fuels as feedstocks.
crop*
HYDROLYSIS
METHANE
Combustion
HYDROCARB
Carton
v-
BIOM ASS-FIRED
POWER PLANT
HB Stock* /-"
/ Methanol (
Synmesis 1
METHANOL \
\Tran»porta(ionFuit-r* | \
Figure 12. Atmospheric sequestration options involving direct-
combustion or conversion utilization of biomass.
1100 ¦'C
50x10* Pt (SO rtn.l
Figure 13. The basic Hydrocarb process involves three
basic process steps which allow conversion of biomass
with fossil fuel to methanol and cartoon.
Direct Combustion or Conversion Utilization
Figure 12 illustrates some of the options for atmo-
spheric C02 fixation via direct combustion or conver-
sion. These options include utilization of atmospheric
carbon fixed as biomass via either direct combustion or
conversion to another fuel form before combustion.
Harvesting new biomass and using it for energy pro-
duction maximizes the benefit from biomass carbon
sequestration since the energy generated can displace
that from fossil sources. Of the various options covered
in Figure 12, direct combustion and the Hydrocarb
process both seem to have special advantages. EPA is
currently pursuing a joint program with the State of
Vermont and Brazil to explore the potential for an
advanced biomass utilization approach involving gas-
ification with turbine combustion/generation. A basic
Hydrocarb process configuration is
shown in Figure 13. It appears to
be a cost-effective option at this:
early point in development with the
potential for leveraging biomass
supply with fossil fuel. Because of
substantial substitution of H, from
biomass, some of the sulfur and ash
free carbon can be used for a power
plant fuel without losing the carbon
sequestration advantage.
Enhancing Uptake by Aquatic
Systems
Figure 14 illustrates some of the
proposed options for enhancing
aquatic fixation, including iron (Fe)
fertilization of microalgae and en-
hanced macroalgae production.
EPA August, 1992
3-8
-------
ATMOSPHERIC C02 (750 GtC)
I (if N or P not kmtting)
!§5f
iria
Microalgae (Phytopiankton)
Cartoon Stored
& 1
Mi]
Ifs-S
J |o il
J't"!
1 |0?
Macroalgae
Caition Stored
PtHSfcta CfttriQ oi
CQ2b*ekiB
Zooptenkton
Ni/trana
Small Amouni
of UttM
€nergy
!CHJ
L
Coral
(epiphytes)
eyeM^ of
COS Mckte
•»no«9«H
CH,. CO,. Sludgo
T
! £
» *
8 I
1 2
2 B
i '
PmM CyCttAQ Of
C02bicfcio
a«K»ph#ie
Marmalyasfarming. Atmospheric
C02 sequestration via macroalgae
farming is illustrated in Figure
15. These ideas are difficult to
evaluate at this point, but Lee
(1991) estimated that to generate
1 x 10" J (1 quad) of natural gas
would require 1000 kelp farms,
each 34 km long and 0.5 km wide.
Capital costs would be over $75
billion. Spencer (1991) has also
estimated mitigation potential
and cost. The following environ-
mental issues must be considered:
Deep Sea (38,000 GtC)
J
Figure 14. Illustration of means tor increasing atmospheric C02 (ixalion via
enhancement of aquatic system uptake. Units are metric gigatons of caroon (GtC)
Ocean fertilization. Microalgae fertilization has been
proposed as a means for enhancing carbon uptake. The
theory behind Fe, manganese, and phosphorus limita-
tions is not well understood or accepted universally.
However, considerable scientific interest in testing the
Fe limitation hypothesis exists, and it will likely be
pursued. Sarmiento (1991) and Joos (1991) have done
3-D modeling simulations of the impact of Fe fertiliza-
tion in the oceans of the Southern Hemisphere. Little
is known about potentially large, adverse effects and
therefore environmental considerations will top the
list of research priorities. Research is likely to focus on
ocean environments where key micronutrients are
thought to be limiting to productiv-
ity. This approach to CO, seques-
tration requires consideration of the
following environmental concerns:
Fertilization of large areas of
the Southern Atlantic and
Southern Pacific Oceans
The potential for deep ocean
• Little is known about environ-
mental effects
• Ocean warming feedbacks
• May require ocean dumping of
processed sludges
• Potential for increased CH4 formation, especially
for sinking options
• Effective use will require C02 disposal which has
several environmental consequences
• Difficulty in supplying and recycling needed
nutrients
• Potential for storm disturbance and loss
• Potential increase in haloform production (methyl
bromide) and increased ozone depletion
• Respiration by organisms forming coral may
release C02 to the atmosphere
Observations and Conclusions
anoxia and feedback of CH4 and
nitrous oxide (NaO)
Potential for increased release
of COj from ocean if fertiliza-
tion stopped
Potential for dramatic changes
in species composition
May favor phytopiankton
species more susceptible to
ultraviolet (UV)-B radiation
Potential for altered fertility in
other ocean regions
Kelp
Farmin
BIOMASS-FIRED
POWER
PLANT
Processing
& Prying
KELP
obile
(¦•¦K1BMI methane
IIBIUIMI'V ENQ"1
1COMPRESSOR
SLUDGE
(Nutrients)
METHANE
C02
sef*rator
(Coral
NUTRIENT
UPWELUNQ
DEVICE
t
C02
to Dissolution
or Deposition
Figure 15, A proposed macroalgae ocean atmospheric CO fixation approach.
EPA August, 1992
-------
Many cost-effective approaches will be needed to deal
with increasing greenhouse gas emissions. Those that
are likely to be implemented before CO, sequestration
are chlorofluorocarbon/halon prevention, prevention
and control of CH4 and other tropospheric ozone pre-
cursors, accelerated conservation, and accelerated
development/use of renewables. Available data on
many CO, sequestration/fixation technology options
are weak, but globally, R&D is increasing in these
areas. Table 3 summarizes the
current picture of the C02 seques-
tration/fixation systems discussed
in this paper. Tables 4 and 5 sum-
marize the overall systems possi-
bilities on a time frame basis. The
underlying construct for these
tables was borrowed from the
Greenhouse Gas R&D Programme
CIEA, 1992) but has been modified
and expanded to cover additional
technologies and atmospheric fixa-
tion options.
posai (for limited quantities) is for product use, includ-
ing chemical feedstocks and enhanced oil recovery.
Many options for co-siting chemical facilities exist
where CO, can be effectively used. While this will not
solve the CO, problem by itself, worthwhile contribu-
tions to reduce CO, can be made. One example that is
already being used is co-siting of ammonia and metha-
nol production facilities. Methanol production opera-
tions can use the extra C03 from the ammonia plant,
and the whole facility can be optimized for energy use.
Table 3. Rue-gas CO, sequestration and atmospheric CO, fixation systems.
COST-
ENVIRONMENTAL
11 fili
TECNNOLOar
MITIGATION
potcktuu.
EFF|CT1VENES$
CONSIOE RATIONS
OIWTTUNmES
Flue Gas Sequestration
Currently, there is no EPA activity
underway on flue-gas CO, seques-
tration, and there is very little U.S.
R&D in this area in general. Glo-
bally, a considerable amount of
R&D work is being initiated: The
Japanese are especially active with
jointgovemment/industry projects,
and other work is underway in Eu-
rope. The major limitations now
are high cost and increased energy
usage. Perhaps some increase in
cost-effectiveness can be achieved
from the simultaneous removal of
SO,, NO,, and particulates along
with the CO,. C0/02 combustion
appears to have research merit as a
technology for concentrating C02,
but much development lies ahead.
1GCC options look especially at-
tractive, but apply primarily to new
power plant capacity. Flue-gas CO,
sequestration requires utilization
or disposal of the C02 in conduction
with any of the above technologies.
The option which offers near-term
market solutions forutilization/dis-
Co-aftrtgaf FCAATQRW
ndum»iC02 m a
p«o«Mfu»i
bMd mw
ognatwro?
fnpm ttytfwfcycal
Mereajgaa oca** LONG TEAM
la*WM»m Papor A mcidMm
atud** UmtfpdMcf
lOH
Many a^nacam
Macwiiipaa tawing LONG TERM
i >nM»f Eartyconoapt
man and EPfllNOAA hava dma
W dwpo—l papor wudiaa
MEDIUM
MEDIUM ro LOW L*y>m»r*»r<*
poi*aaip«M*M.
(a
d
LONG TERM
e
ndu#|
LOW
LOW
OmM&on no—Mi £ii«wiiiii
3-10
EPA August, J 992
-------
Table 4. Overall summary 1Iue-gas sequestration systems.
Ttrn Frame
Technology
COS
Ramoval
COZ Dlepeettlon
Currant
Induitrial proceaaee
(e.g., ammonia 4 methanol
Cacitibee)
Co-iiUnf for
C02 uh
Chemical feeditcci
Naar-ierra
Conventional fouil-rueUd
utility boilara
Ainina *
othar (lue-fai
ramoval tech.
Enhanced oil
recovery ih 4
diapoul in depleted
oil 4 gai formation*
Medium-term
Integrated gaailicauon
combined cycle
COV02 combuabon (new 4
retrofit) with 02 from air
teperation
Absorption,
cry ogemca. 4
membranei
Not required
New chemical
product* and
diipoaal in «alt
domeVaquifere
Long-urn
CC0J02 roml fual cnmbuiuon
with 02 from aotar
photovoltaic eiectrolym of
water
C02/02 'aynthetie
mathanol'-ruiled advanced
comb cyclee (cloaad carbon
eyela)
Othar:
- Other adv. cyclee
- Fual calli
Not required
Not required
Methanol
ttanaportauon and
peaking fual from
C02 4 H2 from
•olar photovoltaic
electrolyaie of waur
- alao provide! 02
for combuction
Ocean diapaaal
Loiil-Urtn
Conventional utility (bull-fuel
combueuon 4 mKiulfia fual
Microti gme
eequeatration
Recycled a» fuel or
uaad for other
product*
Table 5. Overall summary of atmospheric fixation systems.
Time Frame
Technology
C02
Capture
COZ
Disposition
Near-term
Increased teirea trial btomaaa
(via for eat mgmt-. (breatation.
agroforcstiy)
•- Direct combuetian
- Elhenol eonv. hydrolysis
Atm.
fiialiun
None
Medium-term
Increased terrestrial bioma&a
(ahort-rotation intensive culture)
- Advanced cycles
- Hydrocarb
- EUtanol adv. hydroiyna
Atm.
fixation
None
Long-term
Microalga* fertilization (ocean)
Mecroaigae (arming with
ansrobic digestion (ocean)
Aim.
fixation
None
Ocean
disposal
without the problems of C05 disposal; re-
search needs to be done to establish feasibil-
ity and reduce potential costs, especially for
the solar hydrogen production required.
Atmospheric Fixation
EPA and others are actively working on fores-
tation, agroforestry, soil sequestration, direct
combustion/utilization, and the Hydrocarb
process development. Sequestration via mi-
croalgae and macroalgae are long term possi-
bilities requiring years to decades of ecologi-
cal effects research. There is no scientific
consensus on the duration of sequestration or
the effectiveness of biological mechanisms in
carbon fixation over long terms. As we learn
more, more potential problems seem to arise.
R&D on direct utilization of biomass offers
near term benefits. Direct biomass utiliza-
tion is currently utilized in the energy sector
and is projected to be more efficient in the
future. Competition for biomass feedstocks is
a serious limitation. For instance the wood
products industry demands increasing
amounts of fiber and can pay more for the
resource than the energy sector can afford
due to the higher value of*final products. EPA
is currently examining the potential for ad-
vanced cycles using biomass. Development of
cost-effective technologies provides incentives
to grow increasing amounts of biomass. The
Hydrocarb process is a medium-term technol-
ogy and appears to provide means of mitiga-
tion of CO, and other emissions from trans-
portation sources. Current research options
for the terrestrial biosphere research appear
justified; near-term benefits and costs appear
to be reasonable, although the sequestration
potential is limited. The serious environmen-
tal concerns associated with ocean microalgae
fertilization and macroalgae fanning make
them longer-term options requiring intensive
environmental assessment.
Disposal of COz in depleted oil and gas reserves is the
nearest-term straight disposal approach. Ocean CO,
disposal approaches appear to be rather costly and
have a major research need on modeling to determine
the new, longer-term atmospheric concentration levels
resulting from disposal. All of the disposal options are
longer-term options because of the environmental as-
sessment required before implementation. The man-
made carbon cycle has the potential for high mitigation
References
Allam, R. J., and Spilsbury, C., A study of the extrac-
tion of C02 from the flue gas of a 500 MW pulverized
coal-fired power station boiler, presented at the First
International Conference on Carbon Dioxide Removal,
EPA August, 1992
3-11
-------
Amsterdam, the Netherlands (1992).
Arakawa, H., DuBois, J. L., and Sayama, K., Selective
conversion of C02 to methanol by catalytic hydrogena-
tion over promoted copper catalyst, presented at the
First International Conference on Carbon Dioxide Re-
moval, Amsterdam, the Netherlands (1992a).
Arakawa, H., Methods for recycling use of emitted
carbon dioxide: catalytic hydrogenation, Now & Fu-
ture, 7, 10-12 (1992b).
Barchas, R., The Kerr-McGea/ABB Lummus Crest Inc.
technology for recovery of C02 from stack gas, pre-
sented at the First International Conference on Carbon
Dioxide Removal, Amsterdam, the Netherlands (1992).
Brown, L. M., Biological trapping of carbon dioxide,
SERI Microalgal Technology, National Renewable
Energy Laboratory (1990).
Cross, P. J. I., and Goldthorpe, S. H., System studies on
C02 abatement from power plants, presented at the
First International Conference on Carbon Dioxide Re-
moval, Amsterdam, the Netherlands (1992.)
Emmel, T., and Maibodi, M., Retrofit Costs for SO, and
NO, Control Options at 200 Coal-Fired Plants, Volume
I, EPA-600/7-90-02la (NTIS PB91-133322), U. S. En-
vironmental Protection Agency, Research Triangle
Park, NC (1990).
Golomb, D., Herzog, H., Tester, J., White, D., and
Zemba, S., Feasibility, modeling and economics of se-
questering power plant C02 emissions in the deep
ocean, Massachusetts Institute of Technology, Energy
Laboratory, MIT-EL 8&- ,03 (1989).
Hall, D. 0., Woods, J., and House J., Biological systems
for uptake of carbon dioxide, Div. of Biosphere Sci-
ences, Kings College, London, presented at the First
International Conference on Carbon Dioxide Removal,
Amsterdam (1992).
Hendricks, C. A., and Blok, K., Carbon dioxide recovery
using dual gas turbine IGCC plant, presented at the
First International Conference on Carbon Dioxide Re-
moval, Amsterdam, the Netherlands (1992).
Herzog, H., Golomb, D., and Zemba, S., Feasibility,
modeling and economics of sequestering power plant
C02 emissions in the deep ocean, Environmental
Progress, 10, 64-74 (1991).
High, C., and Slog, K, Current and projected wood
energy consumption in the United States, Energy from
Biomass Wastes. XIII, D. L. Klass, Editor, Institute of
Gas Technology, Chicago, IL, pp 229-260 (1990).
Hoffert, M. I., Potter, S. D., Kadiramangalam, M. N.,
and Tubiello, F., Solar power satellites: energy source
for the greenhouse century?, Space Power, 10,131-151
(1991).
Hoffert, M. I., Wey Y. C., Callerari, A. J., and Broecker,
W. S., Atmospheric response to deep-sea injections of
fossil-fuel carbon dioxide, Climatic Change, 2, 53-68
(1979).
IEA, Greenhouse Issues, IEA Greenhouse Gas R&D
Programme, No. 2 (February 1992).
Intergovernmental Panel on Climate Change (IPCC),
Climate Change, The IPCC Scientific Assessment,
World Meterological Organization/United Nations
Environment Programme (1990).
IPCC, 1992 IPCC Supplement, Intergovernmental
Panel on Climate Change, UNEP/WMO (1992).
Joos, F., Sarmiento, J. L., and Siegenthaler, U., Esti-
mates of the effect of southern ocean iron fertilization
on atmospheric COa concentrations, Nature, 349,772-
775(1991).
Lee, K H., Johnston, S. A., Standi, W. D., and Byrd, D.
C., Biomass state-of-the-art assessment, prepared by
Research Triangle Institute for the Electric Power
Research Institute, EPRI GS-7471s Vols. 1 & 2 (1991).
Liro, C. R., Adams, E. E„ and Herzog, H. J., Modelling
the release of C02 in the deep ocean, presented at the
First International Conference on Carbon Dioxide Re-
moval, Amsterdam, the Netherlands (1992.)
Miller, S. L., The nature and occurrence of clathrate
hydrates, In: Natural gases in marine sediments, Ed.
Isaac R. Kaplan, Plenum Press, N.Y., pp. 151-171
(1974).
Moritsuka, H-, Study of power generation systems for
carbon dioxide collection, Central Research Institute of
Electric Power Industry, Yokosuka Research Labora-
tory, EW90006 (1991).
Nishikawa, N., Morishita, M., Uchiyama, M.,
Yamaguchi, F., Ohtsubo, K., Kimuro, H., and Hiraoka,
R., C02 ciathrates formation and its properties in the
simulated deep ocean, presented at the First Interna-
tional Conference on Carbon Dioxide Removal,
**- EPA August, 1992
3-12
-------
Amsterdam (1992).
ORNL,The Carbon Dioxide Information Center, Trends
'90, A compendium of data on global change (1990).
Sampson, R. N., Biomass opportunities in the United
States to mitigate global warming, American Forestry
Association, IGT Biomass Conference (1991).
Wilson, T. R. S., The deep ocean disposal of carbon
dioxide, presented at the First International Confer-
ence on Carbon Dioxide Removal, Amsterdam, the
Netherlands (1992).
Wolsky, A. M., Daniels, E. J., and Jody, B. J., Recover-
ing CO, from large- and medium-sized combustors, J.
Air Waste Manage. Assoc., 41, 449-454 (1991).
Sarmiento, J. L., and Orr, J. C., 3-D simulations of the
impact of southern ocean nutrient depletion on atmo-
spheric C07 and ocean chemistry, Limnology and Ocean-
ography (in press) (1991).
Smelser, S. C., Stock, R. M., and McCleary, G. J.,
Engineering and economic evaluation of CO, removal
from fossil-fuel-fired power plants, Volume 1: pulver-
ized-coal-fired power plants, prepared by Flour Daniel,
Inc., EPRIIE-7365 (1991).
Smith, I. M., and Thambimuth, K. V., Greenhouse
gases, abatement and control, IEA Coal Research,
IEACR/39 (1991).
Sparrow, F. T., Wolsky, A. M., Berry, G. F., Brooks, C.,
Cobb, T. B., Lynch, E. P., Jankowski, D. J., and
Walbridge, E. W., Carbon dioxide from flue gases for
enhanced oil recovery, Argonne National Laboratory,
DE89-014017 (1988).
Spencer, D. F., A preliminary assesment of CO, mitiga-
tion options, The Energy Journal (1991).
Steinberg, M., Lee, J., and Morris, S., An assessment of
CO, greenhouse gas mitigation technologies,
Brookhaven National Laboratory, BNL-46045, (1991).
Steinberg, M., Synthetic carbonaceous fuels and feed-
stocks from oxides of carbon and nuclear power,
Brookhaven National Laboratory, BNL-22785 (1977).
Suda, T., Development of flue gas carbon dioxide recov-
ery technology, presented at the First International
Conference on Carbon Dioxide Removal, Amsterdam,
the Netherlands (1992).
SypherrMueller International, Development of a re-
search, development and demonstration program for
the control of greenhouse gases from coal use, prepared
for Environment Canada (1991).
Vitousek, P. M., Can planted forests counteract in-
creasing atmospheric carbon dioxide?, J. Environ. Qual.,
20, 348-354 (1991). paper (135 been reviewed in accordance wiin the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication.
3-13
EPA August, 1992
-------
The work described in this paper was not funded by the U.S. Environmental Protection Agency. The
contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred.
PAPER 3-C ;
THE NOAA CARBON SEQUESTRATION PROGRAM
Peter Schauffler, Senior Associate
The George Washington University Institute for Urban
Development Research (GWU/IUDR)
Washington, DC 20052
Increasing attention is being given to oceanic techniques for sequestering carbon In
various molecular forms as part of an overall global-change strategy. EPRI, through a
Cattech contract, is looking at the formation and semi-permanent retention of COz
hydrates at ocean depths of 1,000 meters or so. Similar investigations are being actively
pursued by the Japanese. And EPRI and NOAA, assisted by CaKech and George
Washington University, are re-examining the feasibility of large-scale farming of
microaJgae as a way of collecting C02 from the atmosphere and upper ocean layers,
sequestering it in liquid or hydrate form in the deep ocean and/or locking up the carbon
in the farm structure, and using the farm-produced methane and perhaps methanol as a
C02-neutral fossil fuel substitute.
INTRODUCTION
Detailed papers and reports in preparation by EPRI will describe
the CO? hydrate research being conducted at Caltech.*' This work suggests
the liRely feasibility of large-scale hydrate deposition, but the possible
long-term effects on ocean ecology must be examined in detail before any
sound decision on a major program can be made.
Whether or not this technique is determined to be economically and
ecologically acceptable, however, the concept of large-scale ocean farming
as a strategy for significant global-change mitigation should be carefully
investigated. The growing of biomass constitutes a natural and logical
technique for fixing atmospheric carbon in the form of plant hydrocarbons.
To the extent that world-wide fuels are derived from biomass fuels, the
build-up of atmospheric CO2 will be slowed by recycling. And the oceans,
constituting over 70% of the world's surface, may well offer the greatest
opportunity in such a strategy.
* See in particular:
"Climate Alteration - A Global Issue for the Electric Power Industry
in the Twenty-First Century" --
a paper being submitted for publication in the Annual Reviews of
Energy and the hnvironment;
,"Investigations of C02 Hydrate Formation and Dissolution" --
a progress report by the California Institute of Technology.
3-14
-------
1. BACKGROUND
Starting in the late 1960s, the Naval Undersea Center at San Diego (now
the Naval Ocean Sciences Center) initiated some preliminary research and
testing of possible techniques for open-ocean farming of macroalgoe--with
the seaweed product put through anaerobic digestion to produce methane.
The inquiry at that time was driven mainly by a concern about the exhaustion
of natural gas reserves. It was recognized, however, that the plant pressings
and digester residues would also provide important pharmaceuticals and
chemicals (including natural fertilizer and fodder) and thot the farms, as
natural biomes, would ottroct large harvestable fish populations. A small
moored grid (Fig. 1) was installed off the north end of San Clemente Island
and tested briefly with Macrocystis plants tied to the grid lines. And a test-
module design was prepared for an offshore buoy with radial arms and circum-
ferential lines (the plant substrate) —with a pump and tailpipe to bring up
lower-layer water for plant nutrition ( Fig . 2).
With the support of the American Gas Association's Gas Research
Institute, and with modest participation by DOE's Solar Energy Reserach
Institute (now the National Renewable Energy Laboratory), the research and
testing program was taken over by the General Electric Company in the mid
1970s.. A buoy module was constructed and moored in 1500 feet of water
some eight miles off Laguna Beach, again with Macrocystis plants tied to the
lines. Serious problems with plant damage were experienced because of
3-15
-------
divergences between storm-wave frequencies and the buoy-and-mooring
response harmonics, and the.strong current at that location reduced the resi-
dence time and thus nutrient effect of the upwelled water. In spite of these
problems, however, the test demonstrated that Macrocystis plants could be
grown in that setting — and regenerate themselves after suffering storm damage.
A comprehensive summary of this experience is available in
Seaweed Cultivation for Renewable Resources— a compilation of detailed
papers prepared by the'principal program participants, edited by Kimon Bird
ond Peter Benson and published by Elsevier in 1987(1).
The world-wide drop in oil and natural gas prices in the early 1980s
caused this test program to be abandoned. But low-keyed plant research
continued through the decade at Caltech and UCal (Santa Barbara) for .
Macrocystis and at Woods Hole for Sargassum and Gracilaria; and the
anaerobic digestion techniques were pursued at the University of Florida.
Interest in the ocean-farming potentials was rekindled, however, by
"The Hot Summer of 1988." With growing agreement among climatologists
on the link between CO2 emissions and global warming, "tfie greenhouse
effect" started to receive serious public attention. It became clear that all
serious possibilities for long-term control of atmospheric CO2 should be
examined. CO2 is released, of course, in the combustion of all carbon-
based fuels. In contrast with fossil fuels, however, the CO2 from the burning
of biomass fuels is all recovered when the replacement crops are grown.
3-16
-------
Because the oceans cover 70% of the world's surface, they offer the greatest
potential in this respect — if substantial production rates can be achieved.
Added to the current natural absorption in the earth and oceans, this tech-
nique might thus permit an actual reduction in the future atmospheric CO2
concentration.
In recognition of these potentials, and with support from the Electric
Power Research Institute (EPRI), the Board on Biology at the National Academy
of Sciences assembled the principal participants in the 1970s tests for a work-
shop in late 1989 to consider whether a new ocean-farming R&D initiative
was justified. The clear conclusion was affirmative.
This conclusion was reviewed and confirmed in a detailed workshop
conducted by Caltech (again wi th EPRI support) in mid-1990. The implica-
tions were further examined in a NOAA-sponsored seminar and detailed
workshop in 1991; the potentials for specific farm concepts (see below) were
considered in an intensive conference In late 1992 (again with EPRI support);
and the possible arrangements for computer analysis and model design and
testing of these concepts were discussed in a January 1993 workshop at the
David Taylor Model Basin.
3-17
-------
CURRENT PRODUCTIVITY AND COSTS
As indicated above, the key to a successful ocean-farming program
will be the productivity achievable with open-ocean farms. This produc-
tivity will determine the cost of the farm-based fuels and co-products.
The baseline for examining this question is the typical productivity
calculated from past and current growth tests in natural kelp beds here and
abroad.
Benson (2) summarizes two recent-year specific tests as follows:
I n< w Productivity
Researchers Plant Year Location
(DAFMT/VY)
Tseng Laminaria 1987 Yellow Sea 3.8
Neushul & Harger Macrocystis 1990 Ellwood (CA) 6.0
In a paper currently being reviewed for publication, North (3) shows
a range of typical Macrocystis productivities running from about 3.5 to 12.7
DAFMT/AA-
Bird (1), using the Macrocystis growth-rate estimates from the 1970s
GE tests, shows a "baseline" productivity of 13.8 DAFMT/A/Y.
The costs per unit of energy produced from the macroalgal feedstock will
vary greatly depending not only on the seaweed and farm type, funding/
management arrangements and local environment but also on the conversion
process efficiency and the credits obtainable from the marketing of byproducts
and from the avoidance of carbon-emission taxes.
3-18
-------
Using the 1970s GE data, Bird depicted the general relationship of methane
cost to seaweed productivity in a set of curves based on this first set of factors
(that is, without regard to byproduct or tax credits) and depending on the farm
life and maintenance/capital-cost ratio (running from 20 years and 0% [A] to
5 years and 20% [E]) as follows:
UNIT COST
3
CO
£
20-
10-
- PRODUCTIVITY
DAFMT/A/Y
Source: (1)p.337
These figures were calculated with Macrocystis in Southern California
coastal waters, using the conversion efficiencies achieved in the nnid-70s
tests. They could change significantly with a different seaweed, farm concept
and setting, a refined conversion process, and a sophisticated byproduct strategy.
PROJECTED PRODUCTIVITY AND COSTS
The future prospects for increased plant productivity and thus reduced
product cost will depend on a complex interaction of several factors — including
ocean settings, farm geometry, nutrient regimes, species selection, and
3-19
-------
possible genetic engineering. Table 1 lists the principal candidate species
and selection criteria.
These potentials are illustrated by a recent coastal test conducted by
LaPointe and Ryther in the Florida Keys and cited by Ryther (4). In contrast
with the typical figures shown above for Macrocystis and Laminaria, this
Gracilaria test (with favorable current and nutrient conditions) showed a
productivity of 41.2 DAFMT/A/Y. And LaPointe (4) states that productivity
rates approaching these values might also be obtained with some strains of
Sargassum.
Whether productivity approaching these high rates can be achieved in
continuous open-ocean operations will only be determined by further tests.
As one example, such tests might be pointed toward farms consisting of large
grids which would migrate east and west in the Pacific Equatorial Currents
and Countercurrent as shown conceptually in Fig. 3.
The optimum farm configuration and size for such an operation can only
be decided after intensive model testing. One possibility, illustrated in
Fig. 4, would include a servicing vessel to provide low-speed towing for
position-keeping and plant nutrition—with additional nutrients provided by
artificial upwelling and recirculation of the digester liquors through leaker hoses
along the edges of the farm modules. And as suggested in Fig. 5, such farms
might be provided with a submersion capability to improve night-time nutrition
and minimize storm damage to the plants and structure.
3-20
-------
The servicing vessel could have the versatility to vary the product mix
according to world market conditions. The conversion operations involved
here could include fish harvesting and processing in various forms, pharma-
ceutical and chemical production, and (depending on the emissions-tax situa-
tion) carbon sequestration in a CO2 liquid or hydrate form.
To illustrate a further synergy in such an integrated farming operation,
the servicing vessel could include equipment for converting the digester-
produced methane into methanol for delivery to shoreside energy users and
plastics plants, with the tanker backhaul consisting of polyester pellets which
could be shaped aboard ship into the elements (lines, nets, hoses and trusses)
needed for farm expansion and replacement.
The potential impact of byproduct sales on energy costs can be critical .
Wilcox (5) estimated in 1976, for instance, that giving full credit for
byproduct (and particularly fish) sales would reduce the methane cost to
something like $4/MBTU. And the recent increases in plant productivity
cited above would further improve this prospect. Based on the relative prices
of methane and fish products in the mid-70s, Wilcox estimated that devoting
1% of the seaweed crop to fish production would cover 50% of the farm costs.
ISSUES AND RESEARCH NEEDS
The most basic lesson from the 1970s tests — as emphasized by North in
the first of the 1980s reappraisal workshops (6) — is that success in large-scale
ocean farming will depend on a good marriage of the biological and engineerii
3-21
-------
elements. (The biology in the 1970s program was well-conceived, but the
engineering was not.)
The issues relating to this marriage can be arranged on a scale starting
with mainly biological objectives and extending through combined questions
to mainly engineering objectives. This array is shown in Table 2, and the
research needs can be discussed for each numbered issue.
(1) The opportunities for improving productivity have been discussed
above. Since the fuel-product volume will be proportional to the
plant carbon content, a major objective in species selection and
stock refinement (and genetic engineering) will be to maximize
the plant C/N ratio.
(2) The problem of macroalgal releases of methyl halides needs careful
attention as a potential threat to stratospheric ozone. A joint
Caltech/UCal-lrvine research project funded by DOE is studying
the methyl halide releases of natural kelp beds. This work should
be extended to cover other major plant species and to include
possible releases in the plant harvesting and processing stages.
(3) The required nursery techniques will depend on the overall farm
concept and con best be studied as part of an integrated system
analysis.
(4) One example of a favorable ocean-current setting is depicted in
Fig. 3. Detailed analyses should be made of other good possibilities
3-22
-------
around the world — considering Hie current characteristics in terms
of the plant and engineering requirements and the resulting costs
and projected multi-product revenues.
Effective nutrient management (a key factor in issue (1) above) will
require careful cost-and-benefit research to determine the best
nutrient-regime combination for each plant and setting.
Marine-fouling control can be critical in the operation of structured
farms. R&D already performed or in process for naval and commercial
vessels and ocean rigs can suggest the various chemical and mechani-
cal control possibilities and performance prospects (including toxicity
problems), and these will be major factors in the selection of farm
designs. Illustrating the combined biological and engineering
challenge in this program, a biological-control strategy similar to
the shoreside integrated-pest-management technique may turn out to
be part of the marine-fouling solution in large-scale ocean farming.
Further increases in con vers ion-process efficiency are achievable —
again through a combination of engineering (reactor design and
operating conditions) and biology (selection and possible genetic
improvement of the methanogens).
In the design of structured forms, the drag characteristics of the
plants and the farm structure itself will be critical in determining
the overall costs and thus the program feasibility. Computer analysis
3-23
-------
followed by design ond testing of scaled-up models can provide
early indications of this feasibility at modest cost — and this
research must have a high immediate priority.
In the investigation of warm-ocean farming possibilities such as
the Pacific Equatorial Belt depicted in Fig. 3, as indicated above,
the continuous survivability of fragile plants can be critical .
Rather than accepting substantial plant and farm damage during
storm periods, it may be feasible to incorporate in the farm design a
capability to submerge to a plant depth of 50 to 100 feet (below
the zone of intense wave action) and return to the sunlit surface
when the storm has passed. One possible submersion scheme is
shown in Fig. 5, ond there may well be others that offer greater
economy and dependability.
Here again, some initial model testing and then a trade-off
analysis will be required. And the potential benefits in this analysis
may include the ability to improve plant nutrition with routine
night-time submergence.
The selection of materials for the farm elements is principally a
challenge in plastics chemistry ond engineering. The possibility
of a farm thot could largely "grow itself" has been described above.
And major opportunities exist here for plastics recycling — not only
of the farm elements themselves bOt of shoreside plastic wastes
3-24
-------
carried as backhaul (say polyester pellets) in the methanol tankers
and shaped into farm expansion or replacement elements aboard the
servicing vessel. As Cj chemistry is refined in future decades,
furthermore, the potentials for direct shipboard production of
plastics suitable for farm elements may well increase.
This issue again underscores the need to treat the farm program
as an integrated-system challenge — continuously optimizing the
system as a whole for any momentary combination of product markets.
In summary, the overall biological and mechanical farm efficiency and
versatility will be the basic determinant of the farm-based energy costs.
POTENTIAL MARKETS
The overall picture of potential ocean-farm products is shown in Fig. 6.
Included in the page-heading list but not shown in the diagram is the potential
for major cultivation, harvesting, processing and marketing of fish.
To minimize transportation costs and maximize the market-targeting
flexibility, and to justify a major hull for purposes of all-weather survival,
most of the seaweed-to-product conversion processes shown here (plus the fish
operation) should be performed on the servicing vessel.
The existence of major methane pipeline networks in industrial countries
will justify a substantial world-wide delivery of the shipboard digester gas
product in this form by LNG tanker. But because it can be readily produced
from the digester gas aboard the servicing vessel and easily transported by
3-25
-------
conventional bulk ctarrier to a large future global market, methanol will
probably be the main farm product. And these same carriers can transport
ashore the food, pharmaceutical and chemical products shown in Fig. 6.
As indicated above, the product emphasis at any given time can be tailored
to the prevailing mix of price levels in the major world markets.
Because a major ocean-farming program would take several decades to
implement, and because world-wide population growth, lifestyle changes and
environmental constraints over this period cannot be accurately predicted,
it Is impossible to project at this time the size of the market for these various
products. It seems clear, however, that the most important markets will be in
food and energy.
The market for ocean-farm food products (frozen and canned fish, fish
meal, fodder-nourished meats, and carageenan as a meat supplement) will
depend not only on global birth rates and consumption habits but also on
chronic drought and flooding problems (which may be global-warming related),
long-term land use pressures, and a possible decline in agricultural productivity.
The market for ocean-farm energy products will depend on global climate
reactions to the continued build-up of atmospheric CO2 from fossil fuel emissions.
Again depending on population growth and lifestyle changes, the rate of CO2
discharge from the burning of these carbon-based fuels — including the weight
of the oxygen as an integral part of this "greenhouse" gas molecule—will
probably continue to rise above the current level of 22 GTY or 44,000
3-26
-------
tons/minute. And at some point, this huge human intervention in the regime
of the biosphere may possibly trigger one or more of the positive feedbacks
outlined in Table 3 — setting in motion o massive upward spiral .
At that point — unless large-volume fusion power has proven to be safe
and economical — there will be a strong surge of interest in ocean farming as
a protection against runaway global warming. And the corollcry world-wide
demand for farm-derived fuels will constitute a massive market.
The basic questions will then be:
(1) Which of the ocean-farm techniques have the greatest "greenhouse"
control potential? and
(2) How rapidly can they be scaled up?
The answers will depend on the farm characteristics and on the scale of
the crisis. In theory, however, such a program could eventually satisfy the
world's full energy requirements while stabilizing the atmospheric CO2 level .
If world energy consumption were to continue at today's level, for instance,
and if the plant productivity demonstrated in the recent Caribbean coastal
tests can be achieved in open-ocean farms, the total world fuel demand could
be met by the farming of about 5% of the Pacific Equatorial Belt shown in
Fig. 3. This would seem to leave substantial room for the growth of farm-
supplied energy use in subsequent decades.
3-27
-------
One question that comes up in any such theoretical projection is what
would be done about aviation fuel. The world-wide consumption in aircraft,
primarily as kerosene, is only about 5% of the global liquid-fuel consumption —
but as shown in Fig. 7, a British researcher has recently asserted that the NOx
emitted by aircraft turbines at typical cruising altitudes has about thirty times
the infrared-obsorption potency as this same gas at sea level . The "greenhouse"
impcct of aircraft-fuel emissions may thus be a substantial portion of the total
fossil-fuel impact from all sources.
The significance for ocean farming can be seen in two curves (Fig. 8)
resulting from a decompression-chamber test conducted by GE with an aircraft-
turbine combustor in the mid-1980s, using methanol compared with aviation
kerosene (Jet A). For the some power, the combustor radiant heat flux was
much lower with the methanol fuel — resulting in greatly reduced NOx emissions.
What this suggests is that methanol may well be a desirable aviation fuel
in a future "greenhouse-sensitive" world — at least for short- and medium-haul
flights where the two-to-one weight penalty (due to methanol's lower energy
density compared to kerosene) can be tolerated. And for the long-haul flights,
the fuel of choice in this situation could be cryogenic methane from oceans
farms. The extra-potency NOx would still be a problem, but the CC>2 released
by these engines would eventually be completely recovered in the growing of
seaweeds for the replacement fuel .
3-28
-------
CONCLUSION
This review suggests that the ocean-farming techniques may have a very
large potential for production of future fuels and complementary products —
and for stabilizing the global atmosphere in the process. (Coastal versions of
these techniques may well have great utility in absorbing waste nutrients and
recycling heavy metals and other problem-discharge components.)
There ore major questions, however, about the economic and engineering
feasibility of these techniques and about the possible environmental effects.
World-wide concerns about the global-warming impacts of fossil fuels
and about the future adequacy of land-based food production argue for an
intensive R&D program that can answer these questions in the latter half of
this decade.
Only in this way can timely and credible decisions be made about the
proper role of ocean farming in a global strategy for energy and food produc-
tion and environmental protection.
3-29
-------
REFERENCES
1) Seaweed Cultivation and Renewable Resources
edited by Kimon Bird and Peter Benson, Elsevier - 1987
2) Personal memorandum of August 17, 1993
3) Review of Macrocystis Biology
draft chapter by Wheeler North in
Biology of Economic Seaweeds
edited by I Atatsuka., SPB Academic Publishing - (pending)
4) Ocean-Farm Candidate Concept Papers
paper on Coastal Farms by John Ryther
George Washington University, IUDR - November 16-17, 1992, and
statement by Brian LaPointe in post-conference discussion
5) A Conference on Capturing the Sun Through Bioconversion
March 10-11, 1976 Proceedings, p. 271,
published by the Bio-Energy Council.
6) statement by Wheeler North in a workshop on ocean farming,
National Research Council Board on Biology
December 7, 1989, unpublished minutes
3-30
-------
Ft g. l Coastal Grid Farm
Operations and
Processing Plant
Research
Vessel /
Ropes and
sgg. floats form a
iPI Hf ra^ *ee*
HP 11 below the
IIS Wl surface
3-31
-------
Fig. 2. OFFSHORF MODUI F CONtTPT
• KELP PLANTS V • -Kr^^
KELP HARVESTKG SHP
-\
PROCESSUS PIANTS: H0LDIN9 SPACES.
LIYMG QUARTERS, BUOYANCY CONTROL
AND NAVIGATION
HELICOPTER PLATFORM "
: [typ
PLANTS ^ ,7 __N, o*»t nnumi
ICAL| : Wtf 1 1 lU ^ 1 ^>v -
pT- WAVE ACTUATED UPWELLKG PUMP
- " — BUOY
POLYPROPYLENE LINES
SUPPORTING KELP PLANTS
POSITION KEEPING PROPULSOR
STRUCTURAL MEMBERS x -
NUTRIENT DISTRIBUTE SYSTEM) ^
1000 FT. LONG UPWELLINC PIPE
3-32
-------
Fig. 3
PACIFIC EQUATORIAL CURRENT PATTERN
C f "/' Su'"""'
V/K ** c c
' ^CHnUnitnM,
$. Cquatomt Current
CURRENTS
-/ —,/ —!
~WTARCT1CA _^,
hy^TT
-sr—-f»< Oitr ..;
„ ANTARCTICA- ""¦'X
Ech*fl Piolcction
Ocean Areas (Square Miles)
Pacific -
Atlantic -
Indian
Arctic
64,186,300
33,420,000
28,350,500
5,905,700
Total: 131,862,500
3-33
-------
Fig. 4 nCFAN RRTD CONCEPT
PLAN VIEW
Tug/Nursery/Converter
(TNC)
<0
SIDE VIEW
{Gracilaria Mode)
Macroaigae
luo fl
TNC
Oc«m SurfKs * 1 -1-* U
—=—\—i—'*rr
W
-if-
i i i k>ch>H8«5"Pi r 7^"
i > i
UimUlnf ript
u
$
t
t
I
i
Submerged position shown with dashed lines
JJ
¦ * » m m j* * ^*¦ "
U li LA..1I L- J » '' Ui.'J
\
• -V —
Ij-'J
\
3-34
-------
Fig. 5 CONCEPTUAL OCEAN-GRID SUBMERSION SYSTEM
Helium-filled Balloon
ENLARGED SIDE VIEW
(Gracilaria Mode)
Tether
fcjiifblt suppltncntal feiUeor.i for
1 WiC A&df
Towing Direction
Cr«e11irfi-hiryeit1n| fed*
Ocean Surface \
Ltiktr Hen
TWC
A1r Hot*'
Macroalgae
Enlarged Top Valve
rinn *•!»« nm
Buoy
r~—^
luof
Submerged position shown with dashed lines
Sacs- SGZZ «s= SS -=*X ¥331
Buoy
*1r priliari 'iltSpit « nrftci \
(dtp«n4int an plist-growth jtue
and 4
-------
Fig. 6
POTENTIAL OCEAN-FARM PRODUCTS
Fuel Production (eg methane, alcohols. o3t) I PrOt£in (c^. invertebrate and fch cultrvation)
Metallic chemicals
Organles production (eg. jcrtonc. organic adds) I Pharmaceuticals
Cloud nudeation
Biopolymers (e.£ Jgar, ilgm, ora^ce/un)
Fertilizers and soil conditioners
UrPRETREATMENTi!
T
Ci^ + HjO
CARBOHYDRATE;
"RECOVERY
m; 'pressing
BRINE
SUPPLEMENT
l-^JAUHNATECij; ¦
•-EXTRACTION A-
j^'MANNITOU®^ >r
>conversion ^
Na-CO
. ' = ANIMAL.- i
¦t:'yfAsrtES:.Zj
FIBERS
£|?SB
PR
ch4+co2
SLUDGE;3fel
PROCESSING ••
K^OASifS?
.SCRUBBER:
rS HYDROLYSIS;^
^'•RECYCLE;'" *
UQUlDVj&i
NUTRIENTS
(ThETHANsII >;OTH»S i^oTASsDm!
I« -(CH^|»I >v SALTS/} ^CHLORIDE
—-J tmCKc')^
N-fERTILIZER
SLUDGE
r^^-FEED^^f?
" ' SUPPLEMENT.
3-36
-------
THE NEW YORK TIMES, TUESDAY, JANVAFfr' 7, 1992
SCSEAICE
Global Warming Threat
Found in Aircraft Fumes
Nitrogen oxide fumes
emiued by aircraft exert 30
times as great an effect on
climate as the same fumes
emitted at ground level by industrial
processes and the burning of fossil
fuels like coal and oil, British scien-
tists have calculated. ' —^
The nitrogen oxides react with oth-
er chemicals in the air to create
ozone in the troposphere, or lower
atmosphere. At higher altitudes, In
the stratosphere, ozone blocks ultra-
; violet, rays from the. sun. that can
cause cancer in humans. In the tropo-
sphere, ozone is toxic to plants and
animals, and it is oneof a number of
iropospheric gases — carbon dioxide
is the chief one — that trap heat.
Many scientists* believe that in-
creases in the gases as a result of
human activity will cause the earth's
. climate to warm substantially, in
coming decades. - —
If the growth, in aircraft traffic in
the 1990's continues at current rates,
nitrogen oxides would result in the
;trapping, of enough solar energy to
raise the earth's temperature by
about two-hundredths of a degree
Fahrenheit, according to calculations
by Dr. Colin Johnson of the United
Kingdom Atomic Energy Authority's
Harwell Laboratop' and colleagues.
By way of comparison, they calculate
• ed, this is about one-seventh the heat-
ing effect exerted by Increased car-
bon dioxide emissions from 1970 to
1980. o
Working with a computerized mod-
;el of the atmosphere, they found that
nitrogen oxide emissions from air-
craft contributed about as much to
global warming as emissions from
jground-based -sources like automo-
,biles, even though aircraft account
'for only 3 percent of all nitrogen
oxides produced by human activity.
One reason, they said in a paper in the
current issue of the journal Nature, is
that nitrogen oxides last longer when
.they are higher in the atmosphere.
: The heating effect of ozone, carbon
'dioxide and all the other heat-trap-
ping gases may be diminished or en-
hanced by a number of competing
factors, scientists say. Similarly, Dr.
Johnson said, the effect of nitrogen
oxides is only one of a number of
factors that must be considered in
coming to a total understanding of the
climatic effects of ozone. He pointed ,
out, for example, that chlorofluoro- ;
carbons, industrial chemicals that
also trap heat, destroy ozone, cancel- ,
ing out part of the total effect from all
sources. Still, Dr. Johnson said, it is j
necessary to understand the effect of j
nitrogen oxide emissions aloft, since i
aircraft traffic is increasing. |
W/LLMM K. STEVENS !
3-37
-------
Fig. 8
AIRCRAFT TURBINE HEAT AND NOx EMISSIONS
X
3
re
u
5
¦o
IS
400
300
200
100
I
i
i
Takeoff —
7T
]J\ -
Climb
—1
Approach -»-j
A
r |
1
I /
1 1 '
Idle-'
P
\
i
A.
-Tl
~ 1
il ¦
1
1
i l
, 1
,1
U
400
600
800
H
I
l-i
e
w
X
o
z
30
20
10
~T
T
1
Flags Indicate Cruise Points
A
.-A—
400 500 600
Tj, K
700
800
(Courtesy General
Electric Company)
3-38
Hethanol Combustion in a
CF6-80A Engine Combustor
General Electric
Report fR83AE8101
pp. 20 i 26
-------
Table 1 MflCROftl GflE TYPFS AND PROPERTIES
&. fistulosa is float-bearing, arctic
calcareous, widely distributed, small, might
be cultured with other large species
temperate, has float-bearing repro. struct,
subtropic & temperate, one float-bearing sp.
temperate, float-bearing, very durable
tropic, cultivated, mod. size
widely distrib., cultivated, high product,
intensely cultivated, temperate
semi-cultivated, harvested, temperate
temperate, very durable
widely distrib. incl. Sargasso Sea, many spp,
float-bearing, temperate & tropic
SELECTION
CR ITER I AC HIGH PRODUCTIVITY
TOLERATES FULL SUNLIGHT
COPPICEABLE
EASILY HARVESTED MECHANICALLY
CONDUCIVE TOCULTURING AND TRANSPLANTING
REPRODUCTION PROLIFIC
REASONABLY TOUGH
GOOD POTENTIAL FOR BYPRODUCTS
TRANSLOCATES
TOLERATES WIDE ENVIRONMENTAL FLUCTUATIONS
GOOD YIELDS FROM FUEL CONVERSION PROCESS
DISEASE RESISTANT
SUPPORTS AN ASSOCIATED COMMUNITY OF USEFUL ORGANISMS
PERENNIAL AND LONG-LIVED
HIGH SURFACE/VOLUME RATIO
GROWS WELL AT LOW. N-CONTENT
SIMPLE NUTRIENT REQUIREMENTS
HIGH UPTAKE RATES
SURVIVES IN HIGH-ENERGY ENVIRONMENTS
EASILY MOORED OR RESTRAINED
3-39
^1arla
Corallina
Cvstoseira
scKlpni^
Ecrreqia
Eucheuma
Gracilaria
Laminaria
Macrocvstis
Ptervoophora
Saraassum
-------
Table 2
TECHNICAL ISSUES
logy
^ 1). Raising the plant productivity (and maximizing the carbon/nitrogen
ratio) through nutrition management, species selection, stock refine-
ment and genetic engineering.
2) Investigating the extent of methyl halide releases during the candidate-
plant growth and decay phases.
3) Developing economical and dependable shipboard nursery techniques.
4) Selecting the optimum ocean current settings for farm nutrition, storm
minimization, traffic avoidance and market access (see Fig. 3).
5) Achieving an economical and environmentally-acceptable combination
of nutrient recycling, low-speed towing, artificial upwelling and
possible submersion for plant nutrition.
6) Reducing the incidence of marine fouling of the farm structures and
servicing vessels.
7) Increasing the efficiency (cu .ft .methane/kg .solids) of shipboard
anaerobic digestion.
8) Reducing the plant and farm-structure drag to minimize the towing
power requirements.
9) Determining the engineering and economic feasibility of farm sub-
mergence to reduce storm damage and increase nighttime nutrition.
10) Selecting or developing the optimum farm-element plastics —
including the use of recycled plastics and the shipboard shaping of
farm expansion or replacement elements.
ineering
3-40
-------
Table 3
Possible Positive Feedbacks from Global Warming
A. General
1. Increased evaporation persisting as water vapor or low-reflection clouds
2. Increased release of CO2 and possibly CH4 from accelerated aerobic
and anaerobic digestion of dead organisms
3. Possible release of CO2 and CH4 from hydrate deposits
4. Reduced absorption of anthropogenic CO2 due to general heat-induced
reduction of plant growth
5. Reduced snow cover and reflectivity
b. Additional Qcean-Re'ated Possibilities
1. Direct release of dissolved CO2 through increased ocean temperatures
2. Indirect release of dissolved CO2 through violent wave action in
intensified storms
3-41
-------
The Role of DOE Energy Efficiency and Renewable
Energy Programs in Reducing
Greenhouse Gas Emissions
Ca:
I
to
Eric Petersen, Director
Division of Applied Analysis
Office of Conservation and Renewable Energy
U.S. Department of Energy
-------
U.S. Commitment to Global Climate Change Mitigation
• U.S. committed to aggressive, but economically sound, programs to reduce
GHG emissions
• Commitment articulated in U.S. Views on Climate Change
• Major focus of the U.S. approach is energy:
improving the efficiency of energy conversion and consumption
switching away from relatively carbon-intensive fuels
eliminating CFCs in energy-using equipment
-------
Federal Mechanisms to Achieve Energy GHG Reductions
CO
I
Administration comprehensive National Energy Strategy
EPA voluntary conservation programs
DOE programs in energy conservation and energy efficiency,
renewable, nuclear and fossil energy technology R&D
-------
What is CE?
The mission of the Office of
Conservation and Renewable Energy
Develop and promote the adoption of cost-effective renewable
energy and energy efficiency technologies and practices, in conjunction
with the states and with partners in the buildings, industrial,
transportation, and utility sectors, for the benefit of the economic
competitivenessy energy security, and environmental quality of the
Nation.
-------
CONSERVATION AND RENEWABLE ENERGY
Today
\X»
• CE research and development focused on
long-term, high risk activities.
• Office structure not effectively responsive
to needs of the changing market sectors.
Today: • CE program offices reorganized to better interact with
? and respond to the four market end use sectors: Utilities,
Buildings, Industry, and Transportation.
4^
cn
• The Office of Technical and Financial Assistance
created to promote CE technologies and work with the States.
• Strong emphasis is placed on increasing productivity and
enhancing competitiveness.
Development of public/private partnerships is a top
priority.
-------
Options in the Building Sector
SUPPLY
Primary
Fuels
End-Use
Fuels
DEMAND
REDUCTION
Petroleum
i
Natural
Gas
i
I
\ I
Oil, Gas
Coal
Nuclear
Renewables
TXT
Electricity I w^e^eat
rz
Efficient Lighting & Appliances
Advanced Windows & Glazings
Improved Heating/Cooling Equipment
Building Standards/FEMP
y
Heating
Cooling
-------
Quality, Affordable Housing
CO
I
oo
Current Situation
Energy
Average home uses>100 MMBtu/Yr
(> 15 quads total)
60% heating and cooling
23% lighting and appliances
Environment
60 pounds of S02/year/house
3400 pounds of C02/year/house
New homes and appliances
use 100 million lbs CFCs/fr.
Construction wastes
(2.5 tons/house)
Technology
On-site, "Stick" built with
limited quality and high waste
Non-integrated, non-optimized
HVAC, appliances, and controls
Productivity
Construction output/hour
down 1.7%/yr, while manufac.
output up 2.7%/yr
$8 billion/year trade deficit
in building components
Energy costs per house:
$120(VYear
Median new home price rising
50% faster than earnings/Worker
Pii
Mm
M-'vy/vri'tf:
mmm.
4
Initiatives
Century 21 Design,
Manufacturing and Delivery
Improve Shell Technologies
Advanced HVAC/Appliances
Advanced Lighting Technologies
On-Site Power from Fuel
Cells and PV
Energy-Efficient Mortgages/
Home Energy Rating Systems
4
Objectives
Energy
Over 50% reduction in
energy requirements:
Reduction in new home heating
and cooling energy by 2/3
100% efficiency improvement
lighting and appliances
Environment
Carbon emissions from home
energy use down over 50%
Elimination of CFC/HCFCs in
new homes and appliances
Technology
Integrated, optimized
buildings designed and built
in an industrialized system
HVAC load reduced 50%
with advanced windows,
insulation, and design
Productivity
Housing industry
productivity gains
comparable to
manufacturing
Reverse trade deficit:
High-value building products
for export
Household energy costs cut
by 50% in new homes
-------
Options in the Industrial Sector
SUPPLY
Primary
Fuels
End-Use
Fuels
Petroleum
Fuel Oil,
Distillates
DEMAND REDUCTION
Natural
Gas
Coal
Feedstocks
¦ Oil, Gas
Biomass
¦ Coal
& Waste
¦ Nuclear
¦ Renewable
J I
Steam
I Electricity
Efficient Motors
Improved Process Efficiencies
Solid Waste Reductioryutilization
Waste Heat Reductioryutilization
-------
L
Direct Steelmaking
Current Situation
Technology
5.000 Ton/Day Minimum Capacity
Coke Oven/Blast Furnace
Energy
17.1 Million Btu/Ton
Environment
4 b NOx/Ton
2.4 Tons C02/Ton
Coke Ovens Exceed EPA regulations
for SOx and Toxics
Market Size
$46.8 Billion (1% of GNP)
169,000 Employees
11.4% of World Production
Productivity
$135/Ton Iron Production Cost
$470/Ton Avg Steel Production Cost
0.26 Man-hours/Ton Liquid Steel
Environmental Compliance Costs
84 Million Tons/Year
Steel Industry
Initiatives
1,000 Ton/Day Scale-Up
of Successful Pilot Plant
Joint AISI-DOE Direct
Steelmaking Process
Development
Objectives
Technology
1,000 Ton/Day Minimum Capacity
Direct Ironmaking
Coke Ovens Eliminated
Energy
13.5 Million Btu/Ton
Environment
0.0004 lb NOx/Ton
2.16 Tons C02/Ton
No Coke Oven Gases
Market Size
Maintain Share of GNP
No Increase in Employees
Increasing Share of World Production
Productivity
-15% Production Cost Reduction
50% Reduction in Capital Cost/Ton
Reduced Environmental Costs
Sources: Energetics, Inc., "Energy Profiles (or US Industry: Iron and Steel Industry Prolile," December 1990; US Department of Commerce,
Statistical Abstract of the United States 1991; 1989 AISI Statistical Abstract; Paine Webber, World Steel Dynamics service; and Office of
Industrial Technologies estimates.
-------
SUPPLY
Primary
Fuels
End-Use
Fuels
CNG
Coal
Biomass
& Waste
Petroleum
Natural
Gas
Hydrogen
Gasoline
& Diesel
Ethanol
Methanol
Electricity
Oil, Gas
Coal
Nuclear
Renewable
Options in the Transportation Sector
DEMAND REDUCTION
Advanced Internal Combustion
Gas Turbine
Fuel Cells
Electric Drives
-------
c
Advanced Light Duty Vehicles
Current Situation
JSne tgjf
~ 6,095 BtWVehlde Mile
!!Ei|Vl^m^tp$B
« 0.54 Grams/MUe VOC.
j*10.e©mm*flylBeNOx
• i TypfcaJ 3,500pound,
>*§ ^% of WoHel Auto i%mc
liMMarfwt lh:U^#
~ r $344 B8H0f» (6% Of QNP)
• 4,108,000Jobs
Productivity
• 41 e/mlle Operating Cost
• 5.1MilBon BartBl^Day
from Imported Petroleum
Vv-fi :f.jy,
iZtu&f'X-
• BattwyRwwwoh
• Fuel G$M Technologies
• Hybrid Propulsion t;>
Syrten^ ftolfivlu
• Ceramic* In Engirt ^
• Gas Turtt wliliil®
Technology r
• Engine Opfenlzaflon
• Lightweight Materials
• Bioethaira Production
Research
• AMFA
• Afttematfve Fuels Data
Center
r
¦
* D
' -
1
0
v
Objectives
:<> (*IUMMW: •'
• 2£90 Btu/Vehlde Mile
• 100% Alternate Fuels,
: Ethanol Methanol,
Natural Qaa, Hydrogen
Environment
* * Zero Emissions First 60
• Remainder of Range:
: 0.04 Graim/Mlle VOC
0.2 Grams/Mile NOx
v. q Reduction of carbon
5-:] W«mi$sk^ 60-100%';
VeftfctoCharactertsttcs
' • f fc.500 Pound Vehicle Avg,
i Comparable performance
iterlcii' Stoa %$¦
• 90+% of U.S. Auto Marke
i Made In U.S.
,• 30+% of World Auto
' Market Made In U.S.
Productivity
« Comparable Cost
• «100%tOV Fuel from
Domestic Sources
Soutcm: SMMcri AMnd d t«* U.S. 1W1. MVHA Motor Vafttci* FacM • Figure* VI. ORHL Tranaporttikir E/W0y D«J» Book Ed. 12. ConMcd kl OTT. PNU w) EnargaOca
-------
Options in the Utility Sector
SUPPLY
Primary
Fuels
End-Use
Fuels
Nuclear
Coal
Petroleum
Natural
Gas
Renewables
Wind
Solar Thermal
Geothermal
Photovoltaics
Biomas^MSW
Hydropower
Electricity
I
DEMAND
REDUCTION
Advanced Battery/Thermal Storage
Improved T&D Systems
PSM/IRP Strategies
High Temperature Superconductivity
-------
U.S. Electric Generating Capacity Trends
(National Energy Strategy)
900
800
700
600
500
400
GW
NES
Requirements
X
300
New Supply anc|/or
Demand Requirements
Retirements
\
Existing
Capacity
Critical
Decision
Period
1990
1995
2000
2005
2010
-------
CO
I
en
CJl
cr
Electric Utility System
an
^^TrTTT
Generation Transmission Distribution End Use
irm PvMtnn i s'
Utilization (5-10%)
Lftses (&3QH)
iheintwci
tfvesfor
1V>
pswaliWiP
MDentavmentTinriaa^
tf"i Ji
Sources: EEI Electric Utility Statistics (1991). NERC1990 Reliability Assessment (1 (SO),Electric Sales and Revenue 1990 (1992)
EIA Annual Electric Outlook (1991), Electric Light and Power (Apr. 1,19ttt>ual Energy Review 1990 (1991),
EIA Electric Power Monthly (April 1992)
-------
Wind Energy Program
• Market Status
1500 MW of capacity, 2.5 billion kWh
understanding of wind forces and resources
rotor durability issues
• On verge of a major market breakthrough
costs down by two thirds in the last decade to $0.08 per kWh
modular units allow incremental additions with short lead times
zero emissions
? - renewable energy production credit in H.R.776
cn
CT>
• DOE wind energy program major thrusts
development of wind systems with lower costs and greater reliability
with industry collaboration
assisting utilities in evaluation of wind power potential
• Market Potential
1.0 to 2.3 quads in 2010; 2.9 to 10.7 quads in 2030 (SERI 1990)
estimated low-end carbon savings 16 MMT in 2010, 48 MMT in 2030
-------
f
Transferring CE Technology to the Marketplace
CE energy efficiency and renewable energy programs incorporate means to
effectively transfer technologies to the marketplace.
Wide variety of mechanisms used:
• financial partnerships with private industry in technology R&D
• CRADAs between DOE laboratories and private industry
• user facilities for private industry testing of new technologies
• demonstration of new technologies in federal facilities
• interaction with trade and professional associations
• establishment of industry advisory groups
• publications
• education programs
Some mechanisms (CRADAs) used across all programs, others are applied as
appropriate
-------
BUILDING PUBLIC/PRIVATE PARTNERSHIPS
A New Era In Cooperative Leadership
National Technology Initiative
In the fall of 1991, President Bush launched a new Administration-wide initiative to
promote U.S. industry's use of technology to strengthen the domestic economy and
enhance competitiveness in global markets — the National Technology Initiative
(NTI).
As President Bush said, there are
"...steps we can take right now to guarantee progress
and prosperity into the next American Century. We get
there by investing in the technologies of tomorrow, with
Federal support of R&D at record levels. We need to
share the results, get the great ideas generated by
public funds out into the private sector, off the
drawing board and onto store shelves. Our National
Technology Initiative will do just that...."
-------
Building Public/Private Partnerships
A New Era in Cooperative Leadership
(continued)
• Since the launching of NTI in late 1991, CE has entered into about 50
CRADAs, that could lead to the commercialization of energy efficiency and
renewable energy technologies
• Examples CRADAs with significant GHG emissions reduction potential:
cji
CO
waste minimization technology to reduce the use of CFC-based
solvents (Motorola)
waste paper to methanol conversion process (AMOCO)
battery development under the U.S. Advanced Battery Consortium
(Auto industry)
-------
~
Government Industry Battery Cost Sharing R&D
Dollars
In
Millions
240
200
160
120
80
40
0
Highlights
• 14 battery types in 1990 focused to 3 by 1992
• Clean Air Act AmendmentyCalifomia influences
• U.S. Advanced Battery Consortium formed
• Research agenda will be industry driven
Total Battery
R&D
#
1
Battery plant
investment
by industry
Industry fyD
_JL
Government
FVD
_t
1990 1991 1992 1993 1994 1995 1996 1997
-------
LOOKING FORWARD
Strong support for Energy Efficiency and Renewable Energy:
• The FY 1993 funding request of $330 million for energy efficiency R&D is double
the funding for FY 1989 when the Bush Administration took office.
• The FY 1993 funding request of $250 million for Renewable Energy R&D is more
than 65% higher than the FY 1989 funding.
• President Bush established the National Renewable Energy Laboratory in
September 1991. Construction of a $20 million, state-of-the art Solar Energy
Research Facility began in June 1992.
The National Energy Strategy: A Continuing Process
• The NES will continually evolve and be refined as new opportunities and
technological advances emerge.
• Development of NES II is underway.
The work described in this paper was not funded by the U.S. Environmental Protection Agency The
contents do not necessarily reflect the views of the Agency and no official endorsement should be Inferred.
-------
Paper 3-E
FUZZY LOGIC CONTROL OF AC INDUCTION MOTORS
JO REDUCE ENERGY CONSUMPTION
by: R.J. Spiegel and P.J. Chappell
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
J.G.CIeland
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, NC 27709
B.K. Bose
Department of Electrical Engineering
University of Tennessee
Knoxville, TN 37996
ABSTRACT
Fuzzy logic control of electric motors is being investigated under sponsorship
of the U.S. Environmental Protection Agency (EPA) to reduce energy consumption
when motors are operated at less than rated speeds and loads. Electric motors use
60% of the electrical energy generated in the U.S. An improvement of 1% in operating
efficiency of all electric motors could result in savings of 17 x 109 kWh/yr in the U.S.
New techniques are required to extract maximum performance from modern motors.
This paper describes EPA's research program, as well as early stages of work, to
implement fuzzy logic to optimize the efficiency of alternating current (AC) induction
motors.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication.
3-62
-------
2
INTRODUCTION
Electric motors use over 60% of the electrical power generated in the U.S. [1J.
There is a population of approximately 1 billion motors in the country, using over 1700
billion kWh per year. Over 140 million new motors are sold each year. A review of the
U.S. motor population reveals:
90% of the motors are less than 1 hp' (fractional motors) in size, but
use only 10% of the electricity consumed by motors;
95% of the electricity used by motors is consumed by approximately 2%
of the motor population (motors greater than 5 hp); and
• 85% of the electricity used by motors is consumed by less than 1% of
the motor population (motors greater than 20 hp).
Based on these facts, it is clear that large energy savings from improvement in motor
efficiency could be achieved with a relatively small motor population. Each 1%
improvement in motor efficiency could result In:
17 billion kWh per year of electrical energy saved;
* over $1 billion in energy costs saved per year;
an equivalent of 6 -10 million tons" per year of uncombusted coal; and
approximately 15 to 20 million tons less carbon dioxide released into the
atmosphere.
AC induction motors have high reliability and low cost and therefore perform
over 80% of the motor tasks in the U.S. Their speed of operation is determined by
the frequency of the input power, and their efficiency is low when operating at part
load. To control the speed of an AC induction motor and thereby match motor speed
to load requirements requires the use of a device called an adjustable-speed drive
(ASD). Significant energy efficiency gains are achieved when induction motors are
controlled by ASDs [2J. ASDs use semiconductors and switching circuits to vary the
voltage or current and frequency of a motor's power supply thereby controlling the
applied torque and speed to satisfy the process or load requirements [2], ASDs are
•1 hp = 746 We
"1 ton = 907 kg
3-63
-------
3
basically power electronic devices consisting of a rectifier and a computer-controlled
inverter. The rectifier converts the standard 60 Hz AC to direct current (DC). The
inverter then converts the DC output of the rectifier to a variable-frequency, variable-
voltage/current AC.
While ASDs can minimize power losses, they do not optimize operations for
maximum efficiency. The goal of this program is to utilize the inherent capabilities of
fuzzy logic set theory in an integrated intelligent energy optimizer in conjunction with
an ASD to improve the energy or power efficiency of electric motors, primarily AC
induction motors, while at the same time meeting the demands of the process
equipment and load operations which are driven by the motors. Fuzzy logic has the
proven ability to represent complex, ill-defined systems that are difficult or impractical
to model and control by conventional methods [3]. In addition, fuzzy logic is a form
of artificial intelligence that can be implemented in an integrated electronic circuit
device or microchip. This ability is especially important in the case of the modification,
or retrofitting, of existing electric motors, since microchips can be readily added
through an add-on circuit board to existing ASD drives and require little additional
electric power for their operation.
FUZZY LOGIC ENERGY OPTIMIZER
Figure 1 is a block diagram of the overall control approach. A fuzzy logic
energy optimizer is used to control the ASD which in turn controls the motor. A
feedback signal, usually motor speed, from the motor is shown by the dashed line in
the figure to indicate that the control scheme may be open-loop (no feedback) or
closed-loop (feedback).
FUZZY
ADJUSTABLE
LOGIC
SPEED
MOTOR
LOAD
ORTVE
A
*
• motor performance parameters
Figure 1. Fuzzy Logic Energy Optimizer for Improved Energy Efficiency
A motor drive (ASD) may be controlled according to a number of performance
functions, such as input power, speed, torque, airgap flux, stator current, power
factor, and overall calculated motor efficiency. Normally in a drive system, the
3-64
-------
4
machine is operated with the flux maintained at the rated value, or with the voltage to
frequency ratio (V/Hz) held essentially constant in relation to the value at rated
conditions. This allows speed control with the best transient response. The constant
V/Hz approach is used wherever actual shaft speed is not measured; i.e., in open-loop
speed control. The open-loop control approach is most effective when applied to
industry applications that do not require tight or accurate process control such as
pumps or blowers. The computer simulated results presented in this paper are based
on the open-loop control approach.
A fuzzy logic energy optimizer is also being developed for the closed-loop
situation which permits precision speed control. Additionally, work is proceeding on
the development of a fuzzy logic energy optimizer for the most advanced control
scheme known as closed-loop speed control with indirect vector control. This scheme
performs efficiency optimization control without sacrificing transient response. This is
very important for high performance applications such as electric vehicles.
For all the control methods the optimization approach proceeds as follows.
The input power is measured and then the control variables (input voltage, input
current, or input frequency) are varied from the initial setting. The input power is
measured again and compared with the previous value. Based on the sign and
magnitude of the input power signal, as well as the value of the last change in the
control variable, a new value for the control variable is computed using the fuzzy logic
energy optimizer. Sequential decrementation/incrementation is continued until the
minimal input power level is reached. This is the operating point for best efficiency for
the particular load torque and speed condition. If a speed increase is demanded or
the load torque increases, the flux can be established to full value to get the best
transient response. When the new steady-state condition is attained, the fuzzy logic
efficiency optimization search begins again to obtain the most energy efficient
operating point.
Figure 2 is a conceptual flow diagram illustrating details of the fuzzy logic
energy optimizer which is contained within the dotted lines. Detailed explanation for
each block of the fuzzy logic energy optimizer is beyond the scope of this paper. It
suffices to say that the basic underlying principle of operation relies on the fuzzy rule
base consisting of several linguistic IF-THEN rules. A suitable rule base for the open-
loop situation is illustrated below. Additional information on the fuzzy logic concepts
contained in Figure 2 can be found elsewhere [3]. The database includes the
necessary information regarding the motor parameters or other pertinent data. The
"fuzzification" stage is where the process measurements are usually represented as
fuzzy singletons, such as big, medium, and small. The "defuzzification" stage is where
fuzzy outputs are typically converted to real numbers. The most common procedure
for this conversion is the center-of-area method, much like that used for calculating
3-65
-------
5
centroids of various combined areas. The computation unit represents
microprocessor functions for interfacing and recalculation.
I
RULE BASE
DATA BASE
FUZZinCATION
DEFUZZIFICATION
PROCESS/
MOTOR
computation
U NIT
Figure 2. Block Diagram of the Fuzzy Logic Energy Optimizer
SIMULATED PERFORMANCE
The preliminary open-loop fuzzy logic controller [4] was demonstrated by
computer simulation. The control variable was the input voltage. Results show
improvement in motor efficiency using fuzzy logic control while maintaining good
performance in other areas; e.g., maintaining desired torque and speed at steady
levels. For example, Figure 3 compares the efficiency of a motor over a broad range
of loads and operating under both conventional constant WHz control and fuzzy
logic control. The load torque relation to rotor speed simulates the behavior of pump
or fan loads, where load torque is proportional to the square of the rotor speed.
Efficiency improvement by the fuzzy logic energy optimizer was achieved for all
speed/torque combinations.
3-66
-------
6
The following rule base was used for the calculation. Three fuzzy sets (N
standing for negative, P for positive, and Z for zero) were chosen to relate the fuzzy
variables, along with the simple set of rules:
1. IF A Pin IS N AND AV0id IS N, THEN AVna* = N.
2. IF AP,n IS N AND AV0,d IS P, THEN AVn>w . P.
3. IF APln IS P AND AVoW IS N, THEN AV„.W - P.
4. IF APin IS P AND AVoW IS P, THEN AVn„w « N.
5. IF APm IS Z AND AV^ IS ANY, THEN AV„.W = Z.
6
Z
ui
i
20.00 30.00 40.00 SO.00 AO.00 70.00 60.00 90.00 100.00
pttctiHACE lUTce twrt snc#
Figure 3. Fuzzy Logic Control Compared with V/Hz Control
for a 100 HP Motor with Torque Proportional to Speed Squared
Rule 5 is needed for convergence on an optimum input power; i.e., the point where any
small change in voltage results in negligible change in input power. The quantities V0id
and Vn#w represent old and new values, respectively, for the control variable (input
voltage) as the optimization approach proceeds to minimize the input power. The
change in input power level is designated by AP|„. To allow adjustment of step size
(for faster convergence with no overshoot), additional linguistic variables (0.0.,
3-67
-------
7
positive medium, PM, and negative medium, NM) were added. A set of 13 rules was
found to be adequate to relate the variables for the simple control problem.
CONCLUSION
Computer simulated results for the open-loop controller show that the fuzzy
logic energy optimizer can significantly enhance the operational efficiency of AC
motors. Future computer simulation developments will include a closed-loop controller
with a dual-variable (voltage and frequency) fuzzy logic energy optimizer. Additional
effort is taking place to provide fuzzy logic efficiency optimization for induction
motors which use indirect vector control.
Initial results further indicate that fuzzy logic energy optimizers can, in a
collective sense, consistently improve motor operational efficiency over conventional
speed control techniques (ASDs) by increments of 1 to 4%. This is highly significant in
terms of potential U.S. energy savings and pollution abatement possibilities. Figure 4
illustrates potential improvements based on conservative estimates of overage
energy savings for the motor classes indicated on the figure and typical coal-fired
power plant heat rates and emissions. The addition of a fuzzy logic energy optimizer
microchip to a 100 hp motor and ASD should result in energy savings amounting to a
cost payback within 3 to 5 months.
3-68
-------
e
Tons CO, Raducad
par Yaar
jhWh Saved par Yaar
»» Savatf par Yaar
Tona SO, Raduead
par Yaar
A» 1-20 20-20,000
Motors HP HP
Figure 4. Projected Collective Savings from Fuzzy Logic
Motor Control for Improved Efficiency
Once these optimizers have been thoroughly developed using computer
simulation models, prototype hardware devices will be tested in the laboratory. A
block diagram of the motor testing facility is shown in Figure 5. The motor output
power is measured using the dynamometer, and the 3-phase input power is measured
with high precision wattmeters. This configuration allows the motor/drive efficiency to
be determined. A personal computer (PC) microprocessor will monitor the data
acquisition systems and communicate with the ASD to alter the ASD voltage and
frequency output. Various degrees of load on the motor are achieved by varying the
strength of the field in the DC brake via the dynamometer.
3-69
-------
9
*:!
rbi
—*
Microprocessor
(er hmiuuMt* Mjcroetop)
* Effluae) OpWuln Cmi*I
*fmj U*k
* Dili Acqiiacm
* M
-------
The work described in this paper was not funded by the U.S. Environmental Protection Agency. The
contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred.
METHANOL PRODUCTION FROM
WASTE CARBON DIOXIDE
CO
] [
] [
Presented to
Symposium on
Greenhouse Gas Emissions
Mitigation Research
August 1992
By Stefan Unnasch
Acurex Environmental
0830-70101M-1
Acurex
Environmental
CORPORATION
-------
Outline ,
0830-7077M- 2
¦ Motivation
¦ H 2 production
¦ CO 2 production
¦ Methanol production
-------
r«\s
<*yv
KS§§I3fJ
3-73
-------
-------
HYDROGEN PRODUCTION
TECHNOLOGIES
Technology
Methane reforming
GJ
I
"-J
Partial oxidation
Coal gasification
Electrolysis
Thermal dissociation
0830-7077M- 3
Energy Input
Natural gas
Natural gas
Coal
Electricity
Heat
-------
HYDROGEN PRODUCTION COSTS
HoO
OJ
I
-»J
Energy Source
1990 photovoltaic
H2 + 1/2 02
Electricity
($/kWh)
0.101
0830-7077M- 4
Hydrogen
2.4
2000 photovoltaic 0.027 to 0.05 0.6 to 0.9
Nuclear
0.067
1.6
-------
C02 RECOVERY ENERGY
REQUIREMENTS
0830-7077M 5
Energy Requirement
CO2 Source
Technology
(kWhe/lbC02)
Concentrated sources
Ready for use
None
Flue gas
Absorption stripping
0.27
monoethanolamine
'
(MEA)
Flue gas
Solid adsorbents
0.4
Flue gas
O2 combustion/
0 to 0.12
CO 2 recycle
Atmosphere
Various
6 to 10
-------
- C()>CONCENTRATION WITH
OXYGEN COMBUSTION
ESD434-90
CONCENTRATED CO
r
00
NITROGEN
OXYGEN
STEAM
GAS
sm-
lTION^
ft. -J i \ 'j
AIR SEPARATION
OXYGEN
-------
METHANOL PRODUCTION
0830-7077M- 6
u>
i
C02 + 3H2 — CH3OH + H20
CO + 2H2 — CH3OH
-------
3-80
-------
i—— METHANOL PRODUCTION FROM
! ! HYDROGEN AND CO 2
n2, h2o CO 2 recovery
0.37 kWhe
1,375 lb CO 2 RECYCLE I
FLUE j
GAS
DIESEL
HYDROGEN
0.188 (b
ELECTRICITY
SOLAR ENERGY
CONVERSION LOSSES
COAL
0.484 lb MAF
6,765 BtU
ELECTRICITY
0.73 kWhe
METHANOL
-------
CO
I
CD
r\3
-VJ-
•• 'v t •.
METHANOL FROM COAL
FUEL-CYCLE
CONVERT
COAL INTO VEHICLE LOSSES
COAL METHANOL METHANOL 110,900
EXTRACTION, 79 900 TRANSPORT, ^
TRANSPORT ' DISPENSING v- V,
16,500 . 5-000 ' . X
DIESEL t
231,900
TOTAL
ENERGY
INTO
SYSTEM
COAL
METHANOL
VEHICLE POWER 19,600 <17,300 LHV)
-------
- METHANOL from biomass and
! DIGESTER GAS FUEL-CYCLE
J . --- - --
U>
I
00
CO
TOTAL
ENERGY
INTO
SYSTEM
255,000
DIESEL
DIGESTER GAS
H H H
it H *
SEWAGE SLUDGE
CONVERSION
LOSSES
106,000
I METHANOL
f.wsrli TRANSPORT,
DISPENSING
5,000
CARBON BY-PRODUCT 37.6QQ
f
METHANOL
VEHICLE POWER 19,600
s" VEHICLE
LOSSES ;
110,900
-------
ENERGY REQUIREMENTS FOR
METHANOL PRODUCTION
Component
C02 Recovery
From Coal Plant
H2+ Coal-based
Methanol
DH3n- '!7TM- 1fl
H2+ Cone ated
r
Coal (Btu)
6875
6875
—
Power output (kWh)
0.733
0
—
PV equivalent (kWh)
—
0.733
—
CO2 recovered (lb)
1.375
0
1.375
PV input (kWh)
0.37
0
0
H2 electrolysis (lb)
0.188
0.088
0.188
PV input (kWh)
4.0
1.89
3.76
Total PV (kWh)
4.37
2.62
3.76
Methanol (lb)
1.0
1.0
1.0
-------
3-85
-------
COST OF METHANOL
PRODUCTION (1990 $)
Component
C02 Recovery H2+ Coal-based
From Coal Plant Methanol
H2 + Concentrated
C02
Hydroaen
Mass (lb)
Cost ($/lb)
Cost ($)
1.24
0.60 to 2.50
0.74 to 3.10
0.58
0.60 to 2.50
0.35 to 1.45
1.24
0.60 to 2.33
0.74 to 2.89
C02
Mass (lb)
Cost ($/lh)
Cost (' y
9.08
0.02
0.19
Obtained from coal
0.00
0.00
9.08
0.00
0.00
Methanol Production
Capital ($/gal)
15% recovery
Operating ($/gal)
0.64
0.10
0.05
2.34
0.35
0.14
0.64
0.10
0.05
Total Cost ($)
1.08 to 3.44
0.84 to 1.94
0.89 to 3.04
-------
SURELY YOU'RE JOKING
GJ
I
00
0830-7077M- 14
Why generate power from coal if PV is
available?
What are the environmental benefits of
recycling C02?
What are the technical benefits?
Are there cheaper alternatives?
-------
-------
SESSION IV: EMISSIONS AND MITIGATION OF METHANE AND OZONE PRECURSORS
M.J. Shearer, Chairperson
GLOBAL ATMOSPHERIC METHANE:
Trends of Sources, Sinks and Concentrations
by
M.A.K. Khalil, R.A. Rasmussen, and M.J.Shearer
Global Change Research Center
Department of Environmental Science and Engineering
Oregon Graduate Institute
Beaverton, Oregon 97006 U.S.A.
Abstract
The global cycle of methane is driven by emissions around 550 Tg/yr from both
natural and sources related to anthropogenic activities, particularly the production
of food and energy. Major sources are rice agriculture, domestic ruminants, and
wetlands. Methane is removed from the atmosphere mostly by reacting with OH
radicals. Some methane is removed by the soils. Over the past decade methane
concentrations have been increasing at about 1%/yr or 16 ppbv/yr. A record of
atmospheric methane extending back 150,000 years has been constructed from the
analysis of polar ice cores. It shows that, during this time, methane
concentrations have never been more than half of present levels. The recent
increase of methane was probably caused by increasing emissions. Recent changes in
the trend of methane may also be attributed to changing levels of OH. This paper
deals with the changes in global methane concentrations in the past, the causes of
increased levels at present and the future of atmospheric methane. The present
understanding of the global methane budget provides critical facts for policies
related to controlling man-made sources.
4-1
-------
1. INTRODUCTION
It has been nearly 10 years since it was conclusively established that methane
concentrations are increasing in the earth's atmosphere (Rasmussen and Khalil,
1981a, 1981b). Since then, considerable progress has been made towards an
understanding of the global methane cycle and the causes of the increasing trends.
This paper is a review of the current understanding. We examine first the recent
global budgets and then the recent and long-term trends of atmospheric methane.
Next we will review the trends of emissions and the role of human activities.
Finally we will relate the trends of emissions and atmospheric concentrations to
establish a coherent view of the global Cycle of methane and discuss the
implications for the future of methane.
2. GLOBAL BUDGETS
2.1 SOURCES AND SINKS
About a dozen complete budgets of methane have been proposed in recent years, most
of them over the last 10 years (Ehhalt, 1974; Ehhalt and Schmidt, 1978; Donahue,
1979; Sheppard et al. , 1982; Khalil and Rasmussen, 1983; Blake, 1984; Bolle et al. ,
1986; Bingemer and Crutzen, 1987; Cicerone et al., 1988; see also Warneck, 1988).
A number of other studies have concentrated on specific sources and their global
distributions. The published budgets are listed in Table 1 from Khalil and
Rasmussen (1990a). The main sources affected by human activities are rice fields
and ruminants, mostly cattle. Other anthropogenic sources also include biomass
burning, coal mining, oil and gas use, landfills, automobiles, and a variety of
other even smaller sources. The main natural emissions are from the wetlands, with
smaller contributions from a number of other sources, such as the tundra, lakes,
rivers, oceans, and termites.
The total emissions are about 550 tg/yr (1 tg - 10,z gm), of which some 60% of the
emissions are from anthropogenic sources. Once methane gets into the atmosphere it
has a lifetime of 8-10 years. It is removed primarily by reacting with OH radicals
(-490 tg/yr), but smaller amounts (-20 tg/yr) are also removed by the soils, and
even smaller amounts are removed by other chemical processes in the troposphere and
stratosphere. The present imbalance between sources and sinks is about 40 tg/yr,
which is observed as an increasing atmospheric trend of about 1% per year.
2.2 PRESENT GLOBAL DISTRIBUTION
The present global distribution is shown in Figure 1. Because most of the sources
are land-based and many are related to human activities, concentrations of methane
are higher over continents and over the northern hemisphere in general. The present
global average concentration is about 1680 ppbv.
-4-2
-------
Global Distribution of Methane
t OA* lUCOlAC
Minqin
1.59 t 1 ; T I ¦ ¦ I t 1 I 1 1 1 I | — ! ¦ ! ¦ -T —
•1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.6
Sine of Latitude
Figure 1. The latitudinal distribution of methane. Data are average values for
1986 at six long-term flask sampling stations from the arctic circle to
the south pole. The annual average concentrations at Minqin, Gansu,
China are shown for comparison.
3. GLOBAL TRENDS
3.1 TRENDS OF CONCENTRATIONS
The global atmospheric trends are summarized in Figure 2 from Khalil and Rasmussen
(1990b). It shows the results of three systematic global studies spanning the
decade between 1978 and 1988 (Steele et al., 1987; Blake and Rowland, 1988; Khalil
and Rasmussen, 1990b). There is remarkably good agreement among the studies even
though the methods and strategies were quite different.
From the systematic studies, it is apparent that the trends have-not been constant
even over the past decade. The early parts of the record in the late 1970s shows
a faster rate of increase than the more recent measurements in the later part of the
1980s. This variation of the rate of increase has led to some confusion as to how
fast methane is really increasing. Our early work showed rates of increase of
somewhat less than 2%/yr; our later work showed increasing races of 1.4%/yr; Blake
and Rowland (1988) found rates of increase of 1%/yr and Steele et al. (1987)
reported rates of increase of somewhat less than 0.8%/yr. All these findings are
consistent when the variability in the rate of increase is taken into account. This
is shown more clearly in Figure 3 from Khalil and Rasmussen (1990b). Over the
decade the average rate of increase is 1%/yr, although within this decade there were
2-year periods when the rate of increase was as low as 0.7%/yr or as high as 2%/yr
(in the earlier years).
4-3
-------
1700
:660
•8 :620
1580
154C
1500
Khali1 &
Rasmussec
5teele
et al .
BiaJce &
Row!ace
157? 1979 1981 1983 1985 198:
1989
Time ([Months}
Figure 2. Comparison of methane trends from three systematic global long-term
studies (Steele et al., 1987; Blake and Rowland, 1988; Khalil and
Rasmussen, 1990a).
>
CL
a.
X
u
-o
c
CJ
Khali! &
Raseussen
90<& CL
n Blake &
Rowland
Sc eele
ec al .
1982 1983 1984 1985 1986 1987 1988
Time ([Months)
Figure 3. Trends over successive two-year periods and 90% confidence limits in
the middle of the time period of calculation. The rate of increase is
not constant, and trends over short times may not reflect long-term
tendencies.
4-4
-------
Systematic measurements were not taken before this period; however, there are
published data taken during the 1960s and 1970s. An analysis of these data shows
trends of similar magnitude during the 2 decades as shown in Figure 4 (Khalil et
al., 1989a).
Over still longer periods spanning the
last 150,000 years, there is a
remarkable record of atmospheric methane
from the analysis of polar ice cores
(Khalil and Rasmussen, 1989b; Stauffer
et al., 1985; Raynaud et al. , 1988).
This record shows that methane
concentrations have varied naturally
because of changing climatic conditions
from ice ages to inter-glacial periods.
The record also shows that the present
(Figure 5) increase started about only
a hundred years ago with small trends
going as far back as two hundred years.
The pattern of the past rapid increase
was therefore likely to caused by human
activities linked to the rapidly rising
population. During the 150,000-year
record concentrations have never been
more than half of present levels.
1 ISC
i in
iite '
1 208 •
1 ion
¦ 00
too -
4 0 0 -
2 0 0 -
A t>
2000
:80C
J3
a.
CL
c
o
c
«
o
c
o
o
1600 ¦
1400
1200
ioo;
Figure 4.
I960 1964 1968 1972 1976 1980
Time CYrs)
Annual average concentrations
during the 1960s and 1970s
(Khalil ec al.. 1989) (Repro-
duced with permission.)
J
*
.*** •
Jt
6#
a 6 4 t & e& ^
0
-1000
a Khalil I Rumuiiin ( 1882)
O Craig I Chou (1982)
X Old OC (1965-1961)
~ Paarman *1 al. (1988)
A Oaamuaaan ft Khalil (198<)
V Raamuaaan • KhaUl (1986)
¦ Robblna al. (1973)
» Slaullar •< al. (1886)
lie
100
- 400
-201
Time G. P.
Figure 5. The atmospheric concentrations of CH« over the last 1000 years (in
ppbv). (From Khalil and Rasmussen, 1987).
4-5
-------
3.2 TRENDS OF SOURCES
The emissions of methane from domestic ruminants and rice fields have increased
steadily over the past 100 years. More recently, emissions from other human
activities such as oil, gas and coal production, landfills, and other waste
processing have also been increasing. We have compiled data from a number of
sources to estimate the increase of emissions from anthropogenic activities. The
results are summarized in Figure 6.
x
N,
H)
3CC
250
2C C
150 -
1S00 1910 1920 1930 1940 ' 1950 i960 1970 1980
100 -
Figure 6.
¦ Rice
i^Energy Sources
Estimated annual emissions of methane from human-influenced sources.
Domesti c An i mais
Landf i : !s
4. CAUSES OF INCREASING METHANE
The mass balance of methane in the atmosphere can be written as:
i«t)-s(t)-«o
dt t(t)
(1)
where C is the concentration, S the emission rate, and t is the atmospheric
lifetime. The concentration can increase only if the sources increase or the sinks
decrease. For methane, both possibilities are plausible. Since methane is removed
4-6
-------
mostly by reacting with OH, a reduction in OH could lead to an increase of methane.
The decrease of OH was expected because CO and CH4 were both increasing and these
gases are the major sinks of OH. Now it appears that these factors may be
compensated by enhanced production of OH, particularly because of increasing
tropospheric ozone and also because of other considerations. It also appears the
OH concentrations are stabilized against changes (see Pinto & Khalil, 1991; Lu and
Khalil, 1991). On the other hand, there is very good evidence that the sources have
increased, although uncertainties and issues still remain. A good case can be made
that most of the increase of methane has occurred because of increasing emissions.
This still leaves a margin for the contribution of declining OH (Khalil & Rasmussen,
1985, 1987).
5. THE FUTURE
We have shown a coherent picture of the global methane cycle in which the trends
from different experiments agree and the increase both at present and over the past
100 years is dominated by increasing emissions from human activities.
In spite of our present knowledge of the methane budget, there are still too many
unknowns that make it impossible to predict future levels with any degree of
confidence.
If an increase of OH is not the main cause for the slowdown in the methane
concentration then it must be explained by the changes of sources. Increases of
methane over the last century appear to be affected significantly by increasing
cattle populations and acreage under rice agriculture. It is now evident that the
area of rice agriculture is no longer expanding. Higher yields are being achieved
by use of hybrid species and artificial fertilizers neither is likely to increase
methane emissions. Thus the lack of increase in the area of rice planted nay lead
to a lack of increase of methane emissions from this source. Similarly, the world
cattle populations are no longer increasing because of various reasons including
lack of suitable range lands. It seems therefore that the two major sources
affected by human activities, rice agriculture and domestic cattle (and perhaps also
other animals), are no longer increasing because of natural limitations on these
activities. There are a number of more recent sources of methane, mostly related
to energy production and waste disposal such as in landfills, that are probably
still increasing.
In addition to the complications already mentioned, there are two other processes
that may significantly affect future levels of methane. First the draining wetlands
over the past century and the continuing loss of wetlands at present is not fully
understood but may lead to a reduction in natural emissions. And second, there is
the possibility that the warming of the world (caused by increasing C02 and trace
gases) will lead to an increase of methane emissions from natural sources or even
destabilize large reservoirs of methane such as the permafrost, thus overcoming the
diminishing role of the sources we know today. These matters greatly complicate
predicting future methane levels or the benefits of controlling the well known major
sources.
If these ideas are correct two important conclusions emerge:
1). If the sources that contributed to past increases of methane are not increasing
>7
-------
any more, Chen Che pasC record of methane is becoming uncoupled from the future.
Future increases of methane will be caused by newer sources that were not important
in the past thus the past does not serve as an indicator of the future. Population
change for instance is no longer an indicator of future methane emissions. Thus
reliable projections of methane concentrations into the future are virtually
impossible at present. It is even possible that changes in OH, most likely
declining concentrations, will contribute more and more co the future trends of
me thane.
2). The second conclusion is that the global methane trends are slowing down
without any legislative intervention, most likely from natural limits to growth of
the major anthropogenic sources. This aspect is particularly interesting since most
control strategies cannot guarantee the reduction of the growth rate by a factor of
two within a decade as has already occurred. The notion that any trend or an
atmospheric trace gas can continue for decades or centuries has no philosophical or
scientific basis, but is commonly assumed in many assessments of the future levels
of man-made trace gases including methane.
The work described in this paper was not funded by the U.S. Environmental Protection Agency. The
contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred.
ACKNOWLEDGEMENTS
We thank Bob Dalluge for field work in China and Don Stearns for laboratory work.
Funding for this research was provided by grants from the National Science
Foundation (#ATM84-8414020; DPP8717023) and the U.S. Department of Energy (#DE-FC06-
85ER6031). Additional support was provided by the Biospherics Research Corporation
and the Andarz Co. An earlier version of this paper was presented by Prof. M.A.K.
Khalil at the Second International Conference on che Scientific and Policy Issues
Facing All Governments, April 8-11, 1991, Chicago, Illinois, organized by SUPCON
International. The Proceedings of that meeting were published in World Resource
Review.
REFERENCES
1. Bingemer, H.G. , and P.G. Crutzen. The production of methane from solid
wastes. J. Ceophys. Res. 92, 2181-2187, 1987.
2. Blake, D.R. Increasing concentrations of atmospheric methane. Ph.D.
dissertation, University of California at Irvine, U.S.A., 1984.
3. Blake, D.R., and F.S. Rowland. Continuing worldwide increase in tropospheric
methane, 1978 to 1987. Science 239, 1129-1131, 1988.
4. Bolle, H.-J., W. Seiler, and B. Bolin. Other greenhouse gases and aerosols.
IN: The greenhouse effect, climate change and ecosystems, Ch. 4 (SCOPE 29)
(J. Wiley & Sons, N.Y., 1986), pp. 157-198.
4-8
-------
5. Cicerone, R.J., and R.S. Oremland, 1988. Biogeochemical aspects of
atmospheric methane. Global Biogeochemical Cycles 2, 299-327, 1988.
6. Donahue, T.M. The atmospheric methane budget. In: Proceedings of the NATO
Advanced Study Institute on Atmospheric Ozone: Its Variation and Human
Influences. Ed.: A.C. Aikin. U.S. Department of Transportation, Washington,
D.C., 1979.
7. Ehhalt, D.H. The Atmospheric cycle of methane. Tellus 26, 58-70, 1974.
8. Ehhalt, D.H., and U. Schmidt. Sources and sinks of atmospheric methane.
PAGEOPH 116, 452-464, 1978.
9. Khalil, M.A.K., and R.A. Rasraussen. Sources, sinks and seasonal cycles of
atmospheric methane. J. Geophys. Res. 88, 5131-5144, 1983.
10. Khalil, M.A.K., and R.A. Rasmussen. Constraints on the global sources of
methane and an analysis of recent budgets. Tellus &2B, 229-236, 1990a.
11. Khalil, M.A.K., and R.A. Rasmussen. Atmospheric methane: recent global
trends. Environ. Sci. Technol. 24, 549-553, 1990b.
12. Khalil, M.A.K., and R.A. Rasmussen. Causes of increasing atmospheric
methane: depletion of hydroxyl radicals and the rise of emissions. Atoos.
Environ. 19, 397-407, 1985.
13. Khalil, M.A.K., and R.A. Rasmussen. Atmospheric methane: trends over the
last 10,000 years. Acmos. Environ. 21, 2445-2452, 1987.
14. Khalil, M.A.K., R.A. Rasraussen, and M.J. Shearer. Trends of atmospheric
methane during the 1960s and 1970. J. Geophys. Res. 94, 18279-18288, 1989a.
15. Khalil, M.A.K., and R.A. Rasmussen. Climate-induced feedbacks for the global
cycles of methane and nitrous oxide. Tellus 41B, 554-559, 1989b.
16. Lu, Y. and M.A.K. Khalil. Tropospheric OH: Model calculations of spatial,
temporal and secular variations. Chemosphere 23, 397-444, 1991.
17. Pinto, J.P. and M.A.K. Khalil. The stability of OH during ice ages, inter-
glacial epochs and modern times. Tellus 4JS, 347-352, 1991.
18. Rasmussen, R.A., and M.A.K. Khalil. Atmospheric methane (CH«): trends and
seasonal cycles. J. Geophys. Res. 86, 9826-9832, 1981a.
19. Rasmussen, R.A., and M.A.K. Khalil. Increase in the concentration of
atmospheric methane. Atmos. Environ. 15, 883-886, 1981b.
20. Raynaud, D., J. Chappellaz, J.M. Barnola, Y.S. Korotkevich, and C. Lorius.
Climatic and CH« cycle implications of glacial-interglacial CH< change in the
Vostok ice core. Nature 333, 655-657, 1988.
21. Sheppard, J.C., H. Westberg, J.F. Hopper, K. Genesan, and P. Zimmerman.
Inventory of global methane sources and their production rates. J. Geophys.
Res. 87, 1982.
4-9
-------
22. Stauffer, B., G. Fischer, A. Neftel, and H. Oeschger. Increase of atmospheric
mechane recorded in Antarctic ice core. Science 229, 1386-1388, 1985.
23. Steele, L.P., P.J. Fraser, R.A. Rasmussen, M.A.K. Khalil, T.J. Conway, A.J.
. Crawford, R.H. Gammon, K.A. Masarie, and K.W. Thoning. The global
distribution of methane in the troposphere. J. ACmos. Chea. 5, 125-171,
1987.
24. Warneck, P. Chemistry of the Natural Atmosphere. Academic Press, N.Y. ,
1988.
4-10
-------
Paper 4-B
COAL MINE METHANE EMISSIONS AND MITIGATION
by: David A. Kirchgessner
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
Stephen D. Piccot
Science Applications International Corporation
3101 Petty Road
Durham, NC 27707
ABSTRACT
Estimates of methane (CH4) emissions from coal mines range from 25 to 45
Tg/yr with a recent estimate as high as 65 Tg/yr. At 46 Tg/yr, the estimate
produced by this project, coal mines contribute about 10% of anthropogenic CH4
emissions and may contribute significantly to the global change phenomenon. While
emissions from underground mines are now believed to be adequately characterized,
virtually no data are available on emissions from surface mines, and data are totally
lacking on emissions from abandoned/inactive mines and coal handling operations.
The methodology developed to calculate emissions from underground mines is briefly
described, as is the Fourier transform infrared spectroscopy technique being
employed for measuring emissions from surface mines. A nitrogen-flooding technique
for enhancing the recovery of CH4 from coalbeds in advance of mining is described as
a possible measure for mitigating CH4 emissions from underground mines.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication.
4-11
-------
INTRODUCTION
Methane (CH4) is a radiatively important trace gas which accounts for about
18 percent of anthropogenic greenhouse warming. Atmospheric concentrations of
CH4 are now increasing at the rate of 1 percent per year [1]. Although the global
CH4 cycle is not fully understood, significant sources of emissions include wetlands,
ruminants, rice paddies, biomass burning, coal mines, natural gas transmission
facilities, landfills, termites, and tundra [2]. Improved emissions estimates for these
sources will allow their relative contributions to the global CH4 cycle to be better
understood, and will provide a means for focusing future emissions mitigation
research.
Attempts made to estimate global emissions from coal mining operations have
generally relied solely on global coal production data and emission factors derived
from CH4 contents of coalbeds (3,4,5]. These estimates are based on the
assumption that emissions are equal to the amount of CH4 trapped in the coal
removed from the mine. Although this trapped CH4 is liberated when coal is fractured
and removed from the mine, there are other CH4 release mechanisms in the mining
process which this assumption fails to take into account. For example, CH4 may be
released from: (1) exposed coal surfaces throughout the mine workings (e.g., the
roofs, floors, and walls); (2) gas which is trapped in the strata adjacent to the mined
seams; and (3) underlying seams close to the seam being mined. Commonly cited
global mine emissions estimates range from 25 to 45 teragrams {Tg) of CH4/year,
which corresponds to roughly 10 percent of the total annual CH4 emissions from
anthropogenic sources [5]. A recent report contains emissions estimates as high as
33 to 64 Tg CH„/year (6).
Underground, surface, and abandoned or inactive mines comprise the three
general sources of mine related CH4 emissions. Emissions from underground mines
can be liberated from three sources: (1) ventilation shafts; (2) gob wells; and (3)
crushing operations. Ventilation air, although generally containing 1 percent or less
CH4, contributes the majority of mine emissions because of the enormous volume of
air used to ventilate mines. Gob wells are drilled into the area immediately above the
seam being mined. They provide conduits for venting CH4 which accumulates in the
rubble-filled areas formed when the mine roof subsides following longwall mining. Their
purpose is to remove CH4 which would otherwise have to be removed by larger and
more costly shaft ventilation systems. Currently, no published data for the release of
CH4 from gob wells exist. However, preliminary data obtained from the coal mining
industry indicate that gob well CH4 emissions could account for a significant fraction
of the total emissions associated with some longwall mines [7], Emissions data for
crushing operations are also extremely limited.
4-12
-------
In surface mines, the exposed coal face and surface, and in particular areas of
coal rubble created by the blasting operation, are expected to provide the major
sources of CH4. As in underground mines, however, emissions may also be
contributed by the overburden and by underlying strata. Emissions from abandoned
mines may come from unsealed shafts and from vents installed to prevent the buildup
of CH4 in the mines.
The main purpose of this research has been to develop an improved
methodology for estimating global CH4 emissions from underground coal mining
»operations and to produce a global emissions estimate using this methodology where
country-specific estimates are not available. The underground mine methodology
integrates data on coal production, coal properties, coalbed CH4 contents (i.e., the
volume of CH4 per ton of coal), and coal mine ventilation air emissions from U.S.
mines. The objective was to develop a procedure which can be used to estimate mine
emissions from generally available coal analyses and production data where coalbed
CH4 data or emission estimates are not available for a country. This procedure will
be briefly described.
Since emissions data are presently not available for surface mines, the Air and
Energy Engineering Research Laboratory (AEERL) of the U.S. Environmental Protection
Agency (EPA) has embarked upon a measurements program to quantify CH4
emissions from selected surface mines in the United States for later inclusion in this
work. The methodology employed will be discussed. Similarly, virtually no data exist
on emissions from handling operations (i.e., crushing, grinding, transport, and
storage) although their magnitude will certainly depend, to a large extent, on the
desorption characteristics of individual coals. There are also no data available on
abandoned inactive mines; therefore, AEERL is initiating assessments in both of these
categories.
Since one purpose of producing these estimates of emissions is to identify
appropriate targets for control within the coal industry, it is also necessary to
evaluate means of mitigating the emissions. Currently the most logical target for
mitigation is underground mines because they are the largest sources of emissions in
the industry and they consist of one or more point sources. The largest source of
emissions from an underground mine is ventilation air but, because of the enormous
volumes of air produced, CH4 concentrations in the air are typically less than 1
percent. No technologies are currently available to make economic use of such dilute
streams. It is believed that the most effective means of addressing the problem is to
degasify coal seams prior to mining. To make this process more economical the
efficiency can be increased by enhancing the recovery of CH4 from coal. AEERL is
studying a nitrogen-flooding technique developed by the Amoco Production Company
for the coalbed CH4 industry to accomplish this purpose.
4-13
-------
EMISSIONS
Numerous studies have examined the physical relationships which control the
production and release of CH4 by coal. These studies have been conducted either to
evaluate the potential of coalbed CH4 resources or to enhance the safety of
underground mines. Generally, the studies address one of two topics: (1) factors
controlling coalbed CH4 content; or (2) factors controlling the concentration of CH4in
the mine atmosphere and mine ventilation air.
Studies in the first group have identified pressure, coal rank, and moisture
content as important determinants of coalbed CH4 content.' Kim related gas content
to coal temperature and pressure, and in turn to coal depth [8]. After including coal
analyses data to represent rank, Kim produced a diagram relating gas content to
coal depth and rank. Although the validity of the rank relationship has been
questioned, it generally appears to have been accepted by recent authors
[9,10,11,12]. Independently of Kim's work, Basic and Vukic established the
relationship of CH4 content with depth in brown coals and lignite [13].
Several studies have recognized the decrease in CH4 adsorption on coal as
moisture content increases in the lowest moisture regimes [14,15,16]. Moisture
content appears to reach a critical value above which further increases produce no
significant change in CH4 content. Coals studied by Joubert et al. showed critical
values in the range from 1 to 3 percent [16].
Investigations which attempt to identify correlates of CH4 content in coal mine
ventilation air include those by Irani et al. [17] and by Kissel et al. [18]. Irani et al.
developed a linear relationship between CH4 emissions and coal production depth for
mines in five seams. KisseJ et al. demonstrated a linear relationship between CH4
emissions and coalbed CH4 content for six mines. Although both studies suffer from a
paucity of mines and/or seams in their analyses, Kissel et al. made the important
observation that mine emissions greatly exceed the amount expected from an
analysis of coalbed CH4 content alone. Emissions are produced not only by the mined
coal, but also by the coal left behind and by surrounding strata. For the six mines
studied, emissions per ton of coal mined exceeded coalbed CH4 per ton by factors of
from six to nine.
MINE EMISSIONS ESTIMATE
Historically coal mine CH4 emission estimates have relied on coal production and
a value for coalbed CH4 content. The implicit assumption was that emissions were the
same as the CH4 content of the coal removed from the mine. A recent estimate by
4-14
-------
Boyer et al. [6] took into account the fact that emissions are six to nine times that
expected based on coal CH4 content alone. This estimate is considerably more
defensible than earlier ones for this reason. Our study refined the estimation
procedure even further by using a series of regression equations to predict coal mine
CH4 emissions from those coal characteristics known to be related to CH4 content:
depth, moisture content, and indicators of rank such as heating value and fuel ratio
(fixed carbon/volatile matter). The first step in the process uses the above coal
characteristics and one of two regression equations, depending upon heating value, to
produce estimates of in-situ coalbed CH4 content. The coalbed CH4 values along with
coal production statistics are then used in a second regression equation to predict
CH4 emissions. The equations have R2 values from 0.56 to 0.71 suggesting that the
independent variables used explain 56 to 71 percent of the variability in the estimated
values. Although the equations were developed using U.S. coal data, they are believed
to be universally applicable since they employ coal characteristics which are known to
control coalbed CH4 content. The calculations are generally performed at the basin
level of disaggregation since this is the type of coal data usually available. A detailed
description of the estimation methodology is reported by Kirchgessner et al. [19]. It
produces a global estimate of CH4 emissions from underground mines of 36.0
Tg/year. This estimate is believed to be of sufficient quality to obviate the need for
further work on this category of mines.
Very little data exist on which to base estimates of emissions from surface
mines. A single emission analysis has been conducted to date by the EPA at a large
Powder River Basin surface mine in Wyoming [20]. Using open-path Fourier transform
infrared (FTIR) spectroscopy, an emission rate of about 4,814 ma/day was
determined. Using a single coalbed CH4 content for the same county and coal seam,
it was estimated that, at the mine's actual coal production rate of 11.8 million tonnes
per year, potential emissions from the mined coal alone should be 1,008 m3/day. This
would suggest, as noted by Kissell et aJ. [18] for underground mines, that actual mine
emissions exceed, by a factor of about five in this case, the emissions which would be
expected based on coal production and coalbed CH4 content alone.
Rightmire et al. [21], in their study of coalbed CH4 resources in the United
States, report 38 analyses of shallow coals (104 m deep or less) with CH4 contents
ranging from 0.03 to 3.6 m3/tonne coal. One analysis of 9.6 m3/tonne for the
Arkoma Basin was not included because it is known to be anomalously high for
shallow coals. Coalbed CH4 analyses for shallow coals from other countries are
lacking, so this study is temporarily making the gross assumption that the range of
0.03 to 3.6 m3/tonne coal reflects the CH4 content range for shallow coals worldwide.
Multiplying the average value for this range (1 m3/tonne) by 1987 world surface coal
production of about 1.8 x 10& tonnes/year [6], and expanding the results by a factor
of five as discussed above, produces an estimate of about 6.3 Tg/year. Adjusting
4-15
-------
this value upward by 10 percent to represent 1989 coal production yields an estimate
of 6.9 Tg/year. As additional surface mine emissions are sampled under the EPA test
program, the factor by which actual surface mine emissions exceed expected
emissions may change, in which case this portion of the emissions estimate will require
modification.
No data were found on CH4 emissions from handling operations. Boyer et al.
[6] estimate that 25 percent of the CH4 contained in the mined coal is released during
post-mining operations. There is no compelling reason not to follow this precedent for
now; therefore, coal handling emissions were estimated by assuming that 25 percent
of the in-situ CH4 content for all coal produced is released in post-mining operations.
If warranted after further EPA investigation, these assumptions will be adjusted.
Country-specific results of underground mine estimates are shown in Table 1.
TABLE 1. SUMMARY OF ESTIMATED GLOBAL METHANE EMISSIONS FROM COAL MINES FOR 1939
1SB8
Underground
Mm Coal
PrsduOicn
ChM MMhane valun
3
(m /lofwe)
Emissions
Country
t
(10 tonnas)
Disaggregation
Laval
Average
Maximum
Minim um
6 3
{10 m tyr)
(Tg/yr)
China
1.053
21 Provinces
4.0
13.9
2.7
12,942
9.3
Former Soviet Union
418
6 Basins
5.6
9.2
2.2
11,045
7.9
Poland
1 81
3 Basins
7.8
7.8
7.7
5,013
3.6
United States
356
19 Basins
3.9
114
0.2
4,871
3.5
United Kingdom
71'
12 Basins
6.0
18.4
0.3
1,756
1.3
West Germany
73
4 Basins
-
•
1,529
1.1
Australia
59
3 Basins
4.6
7.1
2.1
1.529
1.1
India
95
8 States
2.0
4.7
0.3
935
0.7
South Africa
115
4 Basins
0.9
1.4
0.6
963
0.7
Country Total
2.421
40,583
29.2
Rest ol World
567
9.487
6.8
Total (Underground)
2.988
50,070
36.0
Total (Surfaoe)
9,629
6.9
Total (Handling) '
3,770
2.7
Total
63.469
45.6
*1990-1991 Production
4-16
-------
SURFACE MINE EMISSION MEASUREMENT
A fundamental goal of the sampling methodology is to obtain an emission rate
for.total CH4 emissions from a surface mine. The heterogeneity and size of the
source called for a creative measurement approach. Since smoke releases show that
emissions from surface mines diffuse out of the pit in the direction of the prevailing
winds, a near-ground-Jevel concentration measurement downwind from the mine is
used to estimate a total CH4 emission rate for the mine. A CH4 measurement of the
cross-wind-integrated concentration of the plume at near-ground level is made using
an open-path Fourier transform infrared (FT1R) sensor. Using this near-ground-level
concentration measurement and a measured background or natural- ambient CH4
concentration, the total mine release is estimated using an appropriate plume
dispersion model. If site-specific plume dispersion characteristics can be determined,
they can be used in the model to more accurately represent the behavioral
characteristics of the plume at a given site. Using a tracer gas, these site-specific
plume characteristics can be estimated as described below.
A tracer gas release can be assumed to be a continuously emitting point
source. Based on this assumption and on the results of the smoke release studies
conducted at strip mines,-standard Gaussian dispersion equations can be applied.
When the standard Gaussian equation is integrated across the y direction (y is
assumed to be in the direction normal to the wind direction) from - - to + -, the
following relationship can be developed [22]:
Ccwi « 2Q_ exp [-1/2(H/o2)2] (1)
(2jt)i'2u az
where,
Ccwi = ground-level cross-wind-integrated concentration (g/ms)
Q = emission rate (g/s)
u * average wind speed (m/s)
o2 = vertical dispersion coefficient (m)
H - effective emission height of plume centerline above ground level (m)
For a ground-level source such as a tracer release at a surface coal mine, H is
effectively equal to zero so the exponent of the expression is equal to 1. Thus,
Equation (1) can be simplified to:
4-f7
-------
Ccwi = 2Q
(2tt) 1 /2u az
(2)
Equation (2) can be used to obtain site-specific oz values for a mine if the
values of the remaining unknowns can be determined. Specifically, cz can be
determined for the plume given: (1) a measured tracer gas concentration (Ccwi)
from an FTJR sensor; (2) a measured value of u from a meteorological station located
near the FTIR path; and (3) a known release rate Q from a tracer gas source, such
as a metered gas cylinder located at the mine. To use this technique to estimate
total mine emissions, a number of o2 values must be determined based on tracer gas
releases conducted at several different distances upwind from the monitoring path.
These resulting values are used to construct a relationship of cz versus distance from
the path for the area source. All tracer gas releases used to determine this az
relationship should be conducted as close in time as possible because atmospheric
stability may change, thus changing the az relationship.
A similar and somewhat simpler technique can also be used to assess plume
dispersion characteristics using fewer tracer gas measurements. Given measured
values for the tracer gas release rate Q, tracer release location, wind speed u, and
wind direction, an appropriate area source plume dispersion model can be used to
predict Ccwi tor the tracer gas plume. The model is run to predict concentrations of
the tracer gas at various points along the FTIR monitoring path. These predicted
concentrations are integrated using the trapezoidal rule to calculate a path-
integrated concentration or Ccwi 'or the FTIR monitoring path. The model is run seven
times, once for each of the seven Pasquill-Gifford (P-G) atmospheric stability classes
[22]. These varying P-G assumptions, which incorporate the influence of oz, simulate
increasing atmospheric stability and its effect on the dispersion of the tracer gas
plume. Since several model results are produced, a range of Ccwi values are predicted
under varying degrees of atmospheric stability. The predicted Ccwi value which most
closely matches the Ccwi measured by the FTIR is used to define the P-G atmospheric
stability class which occurred during the tracer gas monitoring event. If simultaneous
CH4 measurements are also collected during this monitoring event, this stability
assumption is applied to the CH4 plume. The model is then run assuming a unity
emission rate for CH4 (i.e., a homogeneous release rate of 1 g/m2-sec) and the P-G
stability determined as described above. The model is run to predict concentrations
of CH4 at various points along the FTIR monitoring path. By again applying the
trapezoidal rule to these predicted point concentrations, a path-integrated
concentration or Ccwi tor the assumed homogeneous release is predicted along the
FTIR monitoring path. Of course the FTIR is actually measuring a path-integrated
concentration due to a heterogeneous emission release pattern from the coal seam.
However, this measured value is comparable to the concentration determined from
'4-18
-------
the model for an assumed homogeneous release because the FTIR measurements
integrated or 'averaged out' the variable concentrations which exist in the plume from
the mine.
The actual CH4 release rate for the mine is then calculated using the simple
relationship shown below where Q(predicted) 'S the unity emission rate for CH4.
Q(aaua/) Concentration(measUred)
= '(3)
Q(predicted) Concentration (predicted)
This technique is used to estimate CH4 release rates in this study. The Point Area and
Line (PAL) source model is used to predict point concentrations along the
measurements path as described above [23]. A non-reactive gas, sulfur hexafluoride
(SF6), is the tracer gas used. Use of a synthetic trace gas such as SF6 is important
to the determination of plume dispersion characteristics because it is non-reactive,
does not naturally occur, and there is no background concentration to cause potential
interferences.
Applying this methodology at a large, strip mine in the Powder River Basin of
Wyoming produced a CH4 emission rate of 4,814 m3/day. An important observation
was, as in underground mines, that actual emissions exceed expected emissions by a
factor of about five in this instance. Details of the methodology have been discussed
previously by Piccot et al. ( 24 ] and Kirchgessner et al. {20]. A validation study of
the methodology designed to answer questions raised during the first sampling trip
has recently been completed and the data are being analyzed.
COAL MINE METHANE MITIGATION
AEERL is participating in a demonstration of the Amoco Production Company's
nitrogen-flooding process to enhance the recovery of CH4 from coal seams. Although
Amoco's interest in developing the technology is focused on CH4 as the saleable
resource, the methods involved will translate fully from the coalbed CH4 industry to
the coal mining industry. The goal of the project is to demonstrate that the 50
percent average CH4 recovery rate from coal seams using current practice can be
increased to 80 percent or more using nitrogen flooding. The final objective of the
Laboratory's involvement is to transfer the practice to the coal industry as needed.
The enhanced recovery, if achieved in a premine degasification program, will allow a
mine to reduce its costly ventilation air requirements, and to retrieve more CH4 for
utilization or sale for a given drilling cost. In this fashion a consistent program of
premine degasification may become not only less costly, but an actual economic
benefit.
4-19
-------
In conventional reservoirs CH4 is contained as a free gas. In contrast, CH4 in
coal seams is stored as a gas adsorbed on the internal micropores of the coal
matrix. The conventional practice of recovering coalbed CH4 is to reduce total
reservoir pressure by pumping water out of the coal. Some CH4 desorbs from the
coal surface, migrates through the micropores to the cleat or fracture system, and
then travels to the recovery well along with the water. Although the system is simple
it is inefficient because at the lower economic limit of pumping, about 150 psi
(lOOOkPa), as much as 50 percent or more of the original CH4 may remain in the coal.
An additional drawback to reducing the total reservoir pressure is that the driving
force for gas expulsion is lost.
An alternative to reducing the total reservoir pressure is to reduce the partial
pressure of CH4 by introducing an inert, low-adsorbing gas at a constant pressure
[25], Partial pressure of a component is equal to the total system pressure multiplied
by the component's mole concentration in the gas phase. Therefore the injection of
nitrogen reduces the relative concentration of CH4 and hence its partial pressure while,
in some cases, increasing total reservoir pressure. Laboratory studies have shown
CH4 recoveries of over 80 percent as well as significantly enhanced rates of recovery.
Modeling studies suggest that the cost of nitrogen is more than offset by the
improvement in production.
The demonstration tract is located in the northern portion of the San Juan
Basin, approximately 9 miles (14.5 km) southeast of Durango, Colorado. The source
of the CH4 is in the coals of the Upper Cretaceous Fruitland formation at a depth of
about 2800 feet (853.5 m). The tract is 80 acres (32.5 hectares) in size with four
injection wells located at the four corners of the tract, and a recovery well located
approximately in the center of the tract. The objective is to demonstrate an
economic CH4 recovery rate of 80 percent or better using nitrogen flooding, with
minimal or no effects on neighboring wells. The project began in the summer of 1992.
SUMMARY
AEERl is actively involved in a program of estimating and measuring global CH4
emissions from coal mines, and of developing mitigation technology for underground
mines should control of this source be deemed prudent. The estimation of emissions
from underground mines is regarded as complete and has produced a value of 36.0
Tg/year. Emissions from surface mines were estimated to be 6.9 Tg/year using a
single measured value. A sampling campaign at selected surface mines will be
conducted using an open-path FTIR instrument and dispersion modeling. If necessary
the estimate for surface mines will be adjusted using these data. Emissions from coal
handling operations were estimated using a technique from the literature, but
4-20
-------
emissions from abandoned/inactive mines have not been addressed. Programs are
underway in both of these areas and the current global estimate of 45.6 Tg/year will
be modified as appropriate when they are complete.
A technology for enhancing the recovery of CH4 from coalbeds is being
demonstrated. It is expected that the increased efficiency will improve the economics
of premine degasification and provide a reasonable method of mitigating CH4 from
underground mines should control of these sources become advisable.
REFERSsCES
1. Smith, B. and Tirpak, D. The Potential Effects of Global Climate Change on the
U.S.: Renort to Congress. EPA/230-05-89-050. U.S. Environmental Protection
Agency, Office of Policy Planning and Evaluation, Washington, DC, 1989. 13 pp.
2. Wuebbles, D.J. and Edmonds, J. Primer on Greenhouse Gases. Lewis Publishers,
Inc., Chelsea, Ml, 1991.
3. Koyama, T. Gaseous Metabolism in Lake Sediment and Paddy Soils and the
Production of Atmospheric Methane and Hydrogen. Journal of Geophysical
Research. 68 (13): 3971, 1963.
4. Marland, G. and Rotty, R.M. Carbon Dioxide Emissions from Fossil Fuels: A
Procedure for Estimation and Results for 1950-1982. Tellus. 36B (4): 232,
1984.
5. Cicerone, R.J. and Oremland, R.S. Biogeochemical Aspects of Atmospheric
Methane. Global Biooeochemical Cycles. 2 (4): 299, 1988.
6. Boyer, C.M., Kelafant, J.R., Kuuskraa, V.A., Manger, K.C., and Kruger, D.
Methane Emissions from Coal Mining: Issues and Opportunities for Reduction.
EPA-400/9-90/008. U.S. Environmental Protection Agency, Office of Air and
Radiation, Washington, DC, 1990. p. 3.
7. S5ot, P. Personal communication. Northwest Fuels Development, Inc., Portland,
OR, 1991.
8. Kim, A.G. Estimating Methane Content of Bituminous Coalbeds from Adsorption
Data. Ri 8245. U.S. Department of the Interior, Bureau of Mines, Pittsburgh,
PA, 1977. pp. 1-22.
4-21
-------
9. Lambert, S.W., Trevits, M.A., and Steidl, P.F. Vertical Bore^le Design and
Completion Practices to Remove Methane Gas from Minable Coalbeds. DOE/CMTC/TF
80/2. U.S. Department of Energy, Washington, DC, 1980. 163 pp.
10. Murray, D.D. Methane From Coalbeds - A Significant Undeveloped Source of
Natural Gas. Colorado School of Mines Research Institute, Golden, CO, 1980.
37 pp.
11. Ameri, S., Al-Sandoon, F.T., and Byrer, C.W. Coalbed Methane Resource
Estimate of the Piceance Basin. DOE/METC/TPR/82-6. U.S. Department of
Energy, Morgantown, WV, 1981. 44 pp.
12. Schwarzer, R.R. and Byrer, C.W. Variation in the Quantity of Methane Adsorbed
bv Selected Coals as a Function of Coal Petrology and Coal Chemistry Final
Report. DE-AC21-80MC14219. U.S. Department of Energy, Morgantown, WV,
1983. pp. 1-6.
13. Basic, A. and Vukic, M. Dependence of Methane Contents in Brown Coal and
Lignite Seams on Depth of Occurrence and Natural Conditions, in: Proceedings
of the 23rd International Conference of Safety in Mines Research Institutes.
U.S. Department of the Interior, Bureau of Mines, Washington, DC, 1989. pp.
282-288.
14. Anderson, R.B. and Hofer, L.J.E. Activation Energy of Diffusion of Gases into
Porous Solids. Fuel. 44 (4): 303, 1965.
15. Jolly, D.C., Morris, L.H., and Hinsely, F.B. An Investigation into the Relationship
Between the Methane Sorption Capacity of Coal and Gas Pressure. The Mining
Engineer. 127 (94): 539, 1968.
16. Joubert, J.I., Grein, C.T., and Bienstock, D. Effect of Moisture on the Methane
Capacity of American Coals. Fuel. 53 (3): 186, 1974.
17. Irani, M.C., Thimons, E.D., and Bobick, T.G. Methane Emission from U.S. Coal
Mines, a Survey. IC 8558. U.S. Department of the Interior, Bureau of Mines,
Pittsburgh, PA, 1972. pp. 7-15.
18. Kissel, F.N., McCulloch, C.M., and Elder, C.H. The Direct Method of Determining
Methane Content of Coalbeds for Ventilation Design. Rl 7767. U.S. Department
of the Interior, Bureau of Mines, Pittsburgh, PA, 1973. pp. 1-9.
i
19. Kirchgessner, D.A., Piccot, S.D., and Winkler, J.D. Estimate of Global Methane
Emissions from Coal Mines. Chernosphere. (In Press), 1992.
4-22
-------
20. Kirchgessner, D.A., Piccot, S.D., and Chadha, A. Estimation of Methane
Emissions from a Surface Coal Mine Using Open-Path FTIR Spectroscopy and
Modeling Techniques. Chemosphere. (In Press), 1992.
21. Rightmire, C.T., Eddy, G.E., and Kirr, J.N. Coalbed Methane Resources of the
United States. AAPG Studies in Geology Series #17. American Association of
Petroleum Geologists, Tulsa, OK, 1984. 9 pp.
22. Turner, D.B. Workbook of Atmospheric Dispersion Estimates. EPA Report AP-
26 (NTIS PB191482). U.S. Environmental Protection Agency, Research Triangle
Park, NC, 1970.
23. Petersen, W.B. and Rumsey, E.D. User's Guide for PAL 2.0 - A Gaussian Plume
Algorithm for Point. Area, and Line Sources. EPA-600/8-87-009 (NTIS PB87-
168787). U.S. Environmental Protection Agency, Atmospheric Sciences
Research Laboratory, Research Triangle Park, NC, 1987.
24. Piccot, S.D., Chadha, A., Kirchgessner, D.A., Kagann, R., Czerniawski, M.J., and
Minnich, T. Measurement of Methane Emissions in the Plume of a Large Surface
Coal Mine Using Open-Path FTIR Spectroscopy. ]q: Proceedings of the 1991 Air
and Waste Management Association Conference, Vancouver, B.C., 1991.
25. Puri, R. and Yee, D. Enhanced Coalbed Methane Recovery. Presented at: 65th
Annual Technical Conference and Exhibition of the Society of Petroleum
Engineers, New Orleans, LA, September 23-26, 1990.
4-23
-------
Emissions and Mitigation
of Methane
from the Natural Gas Industry
Robert A. Lott
Gas Research Institute
8600 W. Bryn Mawr Avenue
Chicago, Illinois 60631
ABSTRACT
The Gas Research Institute and the U.S. Environmental Protection Agency are
cofunding and comanaging a program to evaluate methane emissions from U.S. natural
gas operations. The purpose of the program is to provide an emissions inventory
accurate enough for global climate modeling and for addressing the policy question
of "whether encouraging the increased use of natural gas is a viable strategy for
reducing the U.S. contribution to global warming.11 The program is composed of
three phases: Scoping, Methods Development, and Implementation.
The purpose of Phase I was to define the problem. Phase II of the program
concentrated on developing techniques for measuring steady or fugitive emissions
and for calculating the highly variable unsteady emissions from the variety of
sources that comprise the gas industry. Because of the large number of sources
within each source type, techniques were also developed for extrapolating
emissions data to similar sources within the industry.
Phase III of the program was started in early 1992 and should be completed in
early 1994. The purpose of the current phase of the program is to collect
sufficient data to achieve the accuracy goal of determining emissions to within .
± 0.5 percent of production.
Based on the limited amount of data collected to date, methane emissions from the
U.S. gas industry appear to be in the range of 1 percent of production.
4-24
-------
SECTION 1.0
INTRODUCTION
The combustion of natural gas produces less carbon dicxide per unit of energy
generated than either oil or coal. For this reason, a number of organizations
have suggested that global warming could be reduced by encouraging fuel switching.
However, methane is a more potent greenhouse gas than carbon dioxide. Since
natural gas is approximately 90 percent methane, leakage of natural gas could
reduce or even eliminate the inherent advantage that natural gas has because of
its lower carbon dioxide emissions.
In order to accurately evaluate the impact of various fuels, the emissions of all
greenhouse gases must be considered as well as the end use efficiency. In such
an analysis, two major issues have been identified: The magnitude of the methane
emissions for each fuel (coal, gas, and oil), and the impact of methane relative
to carbon dioxide; i.e., the global warming potential (GWP) of methane. The
uncertainty surrounding these two issues overwhelms all other considerations. For
this reason, the Gas Research Institute (GRI) and the U.S. Environmental
Protection Agency (EPA) have developed a jointly funded and managed program to
better define emissions from the U.S. gas industry. It is a comprehensive study
to quantify methane emissions from the wellhead to and including the customer's
meter.
The goal of the study is to determine emissions to within 0.5 percent of
production or approximately 100 billion cubic feet (BCF)* per year.
The purpose of this paper is to describe the GRI/EPA program, present interim
results, describe future work, and discuss mitigation strategies. These four
topics also comprise the four major subsections of the paper.
* English units are used in this paper because it is the accepted practice in the
U.S. gas industry. The table below cam be used to convert from English to
metric units.
Conversion Table
FROM
TO
MULTIPLY BY
ft
m
0.304B
ft3
m5
0.02832
lb
kg
0.4536
hp
watts
745.7
mile
km
1.610
4-25
-------
SECTION 2.0
DISCUSSION
2 .1 GRI/EPA Program
The GRI/EPA program is divided into three phases:
Phase I Scoping
Phase II Methods Development
Phase III implementation
2.1.1 Phase I
Scoping studies were conducted by GRI and EPA independently, before
the formation of the joint program. It was clear from these
scoping studies that three major problems needed to be solved in
order to quantify methane emissions from the U.S. gas industry.
First it was recognized that all types of emissions could not be
measured. Steady emissions -- which are defined as emissions that
are nearly constant in time such that a measurement made over a
relatively short period (1 hour) is representative of the annual
average value -- can be evaluated by measurement. However, it
would be impossible, certainly impractical, to try to measure
unsteady emissions; i.e. , those that are highly variable with time.
Therefore, techniques needed to be developed to calculate the
unsteady emissions from all the different source types.
The second problem was that proven techniques for measuring the
steady emissions from the different source types were not
available. New techniques needed to be developed and validated.
Lastly, it was clear that, even if methods for calculating the
unsteady emissions and measuring the steady emissions were
available the emissions from all sources could not be evaluated
because the number of sources was overwhelming. For example, there
are a quarter million gas wells, over a million miles of pipe, and
hundreds of thousands of pressure regulators. Because emissions
from all these sources could not be measured, scientifically
defensible techniques needed to be developed that would allow data
obtained for a set of sources to be extrapolated to similar sources
throughout the industry.
In summary, the scoping studies established that three major tasks
needed to be accomplished in Phase II or the Methods Development
Phase of the program. These are to develop methods for:
0 Measuring Steady Emissions
° Calculating Unsteady Emissions
0 Extrapolating Emissions Data
4-26
-------
2.1.2 Phase II
The block diagram in Figure 1 illustrates the approach taken in the program.
The gas industry was disaggregated until techniques could be identified for
evaluating emissions from the various types of sources. In some cases
emissions from the basic components of the industry {e.g., pipes, valves,
flanges) needed to be considered. It was necessary to develop techniques not
only for measuring the steady emissions from these sources, but also for
calculating the unsteady emissions. In addition, accuracy targets for the
emissions data from each source had to be established.
Accuracy targets were established in such a way that, if the target is met
for each source, the overall accuracy of the program (+ 100 BCF) would
automatically be achieved. In addition, the accuracy target is established
as a function of the size of the source. A higher degree of accuracy is
required for the larger sources, and this approach provides an automatic
mechanism for apportioning the needed resources to the most important
sources.
The flow chart for Phase II is presented in Figure 2. The first step is to
gather information on the gas industry in order to better define and
understand the problem to be solved. Developing methods for measuring,
calculating, and extrapolating emissions proceeded along parallel tracks.
The last step of the process is to validate the methods developed. The
experimental methods were validated by conducting a "proof of concept" test
not only to demonstrate that the method worked under controlled conditions,
but also to establish error bounds. The technique was then demonstrated in
the field by measuring emissions from actual sources. The methods for
calculating the unsteady emissions and for extrapolating the data were
validated by documenting the methods and subjecting them to a critical review
by experts.
Measurement Techniques
Developing techniques for measuring steady emissions from all the different
types of sources has presented the most difficulty. Figure 3 lists the five
techniques that have been developed for measuring emissions from the
different segments and source types comprising the industry. The essence of
each method is described below.
Emission Factor Approach This approach is used to determine emissions from
different source types based on measurements of emissions from individual
pipe fittings such as valves, flanges, seals, and threaded fittings.
Emissions from a large number of fittings are measured and an emission factor
(i.e., average emission rate per fitting) is determined for each fitting
type. Emissions from a source are calculated by multiplying the number of
fittings comprising the source by the appropriate emission factor. The
emissions are measured using the bagging technique; i.e., enclosing the leak
in a bag, blowing uncontaminated air through the bag at a constant rate, and
measuring the concentration in the stream. The emission rate is the product
of the flow rate and concentration.
4-27
-------
Correlation Equation Technique Like the Emission Factor Approach, this
technique is based on measuring emissions for individual pipe fittings.
However, a screening value is also obtained. The screening value, which is
the maximum concentration measured on the fitting at the point of the leak,
is correlated with the emission Tneasurement. To determine the emissions from
a specific source, the concentration measured from every leaking fitting is
used with the correlation equation to calculate the emission rate for each
fitting. The emissions from all the fittings comprising the source are
summed to obtain the total value.
Tracer Gas In this approach,.a tracer gas is released at a known constant
rate at the methane source, and the concentration of both the tracer gas and
the methane is measured at a point downwind where these gases are uniformly
mixed. The emission rate of the methane source (EM) is calculated from the
expression EM = (CM/CT)ET, where CM is the methane concentration corrected for
background, CT is the tracer gas concentration, and ET is the emission rate
of the tracer gas.
Leak Statistics Method This method is directed toward evaluating emissions
from buried pipelines in distribution systems and gathering lines in
production fields. Emission rates are measured for a large number of leaks
in order to accurately determine the average emission rate per leak as a
function of pipe material, age, pressure, and soil characteristics.
The leak data recorded by individual gas distribution companies are
statistically analyzed to determine the actual number of leaks in the system.
Total emissions from the underground pipe system are calculated by
multiplying the appropriate average emission rate per leak by the number of
leaks per mile and the number of miles of pipe in each category.
Mass Balance Approach This method uses the existing metering system to
perform a mass balance. The difference between the amounts entering and
leaving a system is equal to the amount leaked to the atmosphere plus an
error term. The key to the technique is to minimize the error. This can be
accomplished by carefully selecting the site and by calibrating and
characterizing the meters in the system.
Calculating Unsteady Emissions
Techniques for calculating the unsteady emissions are described in Reference
1.
Emissions from a given source can be caused by any of the following
activities:
Normal operations.
Maintenance,
Upsets,
Mishaps,
Leaks, and
Combustion.
4-28
-------
Because the technique for calculating the unsteady emissions is dependent on
not only the type of source, but also the cause of the emissions, as many as
six equations are needed to specify the unsteady emissions from a given
source type.
Initial studies indicated that emissions from compressor exhaust and venting
activities accounted for nearly 98 percent of the unsteady emissions.
Therefore, Radian Corporation, the contractor for this portion of the work,
focused their resources during Phase II on these two categories.
Extrapolation Technique
Emission and activity factors will be used to extrapolate the measured or
calculated emissions data. These factors are defined in such a way that
their product will be equal to the total emissions from a given source type
or category.
In the simplest case, the emission factor would be defined as the average
emission rate calculated from a large number of measurements of randomly
selected sources of a given type. The activity factor would then be the
number of sources within the U.S. For example, total U.S. emissions from
pressure regulating and metering stations could be calculated by multiplying
the emission factor, which would be the average emission rate (cubic
feet/station-year), by the activity factor, which would be the total number
of stations.
In applying the approach, two major problems can be encountered. The first
is that the variability in the emissions data is very large which could mean
a very large number of expensive tests would be required to achieve the
desired accuracy target. This problem can often be alleviated by presenting
the emission and activity factors as a function of other parameters that
affect the emission rate.
The second problem is that the sources tested cannot be randomly selected for
a number of practical reasons. This can induce bias in the results. The
bias can often be found through an analysis of the data. For example, Table
1 presents the measured emission rate from 39 pressure regulating and
metering stations along with the emission and activity factors and the
extrapolated emissions for this source type. The extrapolated emissions for
this source category are 140 BCF. However, if the data are subdivided into
two categories (pressure regulating stations, and regulating and metering
stations), the estimated emissions for this category decrease to 81 BCF. If
these categories are then subdivided into discrete operating pressure ranges,
the estimated emissions decrease to less than 27 BCF. The bias, which was
caused by testing a disproportionate number of high pressure facilities, can
be removed by refining the emission factor. As shown in Table 1, the
extrapolated emissions decreased by over a factor 5 as a result of analyzing
the data to refine the emission and activity factors.
4-29
-------
2.2 Results
2.2.1 Measurement Techniques
As part of the process to develop and validate experimental techniques, tests
were conducted to first prove the concept. These tests were then followed
by a series of demonstration tests to verify that the technique could be used
to measure emissions from specific types or groups of sources. The purpose
of the Phase II effort was to develop and validate methods, not to collect
data on a wholesale basis.
Proof of Concept Tests
Emissions Factor/Correlation Equation Method These methods are standard
techniques approved by EPA (Reference 1) for measuring fugitive or steady
emissions from oil, gas, and chemical production and processing facilities.
The methods were originally developed based on data collected in the late
1970's. Because the validity of these data has been questioned, tests were
conducted in the spring of 1990 at a gas/oil production site. Although the
test program was limited, it was determined that the emission factor approach
over estimated emissions by a factor of 5 to 20, that there was large
uncertainty in the correlation equation approach, and that an extensive data
collection effort was required to develop new emission factors and to select
the proper correlation equation from the many found in the literature
(Reference 2).
Tracer/Proof of Concept Test A proof of concept test was conducted by SRI
International to demonstrate that a tracer gas could be used to measure
emissions from a point source. In this test, methane, ethane, and a tracer
gas were released at a constant rate from the same location. The release
rate was carefully measured for all the gases. Air samples were collected
at 15 minute intervals at three downwind distances and analyzed to determine
the concentration of the gases. It was found that the release rate of methane
and ethane could be calculated to within 10 percent of the true value
(Reference 3) using the relationship that the ratio of emission rates (Em/Et)
is equal to the ratio of concentrations (CM/CT) .
A similar proof of concept test was conducted by Washington State University
(WSD) teamed with Aerodyne Research Inc. (ARI). For this test a real-time
instrument, developed by WSU for measuring the tracer gas, was mounted in a
truck with a helium/neon laser, developed by ARI for measuring methane. The
laser system is capable of measuring methane in real-time to within + 5 ppb.
These mobile instruments were then used to traverse the plume generated by
releasing methane and the tracer gas at a known constant rate. Tests were
conducted with the sources colocated and also separated by 100 feet. In both
cases, the emission rate of the methane source was calculated to be within
10 percent of the true value (Reference 4).
Leak Statistics and Mass Balance These techniques have been used by the gas
industry to account for the gas moving through the system (Reference 5) . The
difference between the metered volumes entering and leaving the system is
equal to that used in the system plus a residual often called "unaccounted
4-30
-------
for" (UAF) gas. UAF gas results from accounting errors, metering errors,
theft, and leakage. To assess the leakage component, the industry has used
the leak statistics method for a number of years; therefore, there was not
a need to perform proof of concept tests for either of these methods.
Demonstration and Other Test Results
Tracer Method
SRI APPROACH... SRI International conducted tracer tests at two gas/oil
production facilities. The test results are documented in Reference 5. The
purpose of these tests was to demonstrate that the tracer technique could be
used to measure emissions from sources generally found at
production/processing sites such as wells, separators, heaters, gas plants,
and compressor stations. SRI used as many as 30 sequential samplers to
collect air samples over eight 15 minute periods. The samples were later
analyzed in a field laboratory using a gas chromatograph to measure the ratio
of methane to tracer gas. Although the tests were successful, it was
concluded that real-time instruments were needed if tracer tests were to be
conducted routinely. Analysis of the large number of samples (20 to 30
samplers with 8 bags per sampler) was too expensive. Also, it was hard to
detect interference from other sources, and the success or failure of the
experiment could not be confirmed until all the air samples had been
analyzed. It was also determined that the tracer technique could not be used
effectively to measure the emission rate from wells because only 2 or 3
percent leak at a detectable rate and significant resources are needed merely
to find the few leaking wells.
ARI/WSU APPROACH. . . A team consisting of ARI and WSU personnel demonstrated
that the tracer or modified tracer technique could be used to measure
emissions from production/processing facilities and from distribution
systems. The team used a truck outfitted with instruments capable of
measuring methane, SF6 (tracer gas], and carbon dioxide nearly continuously
in real time.
These tests (References 6 and 7) demonstrated that the tracer technique is
well suited for measuring emissions from a point source or any group of
spatially concentrated sources. These include metering/pressure regulating
facilities, compressor stations, gas processing plants, separators, and
metering facilities. The method is not well suited to measure emissions from
wells and customer meters because of the large effort required to identify
the few that leak. Tracer tests of 39 pressure regulating/metering stations
are summarized in Table 1 and were discussed in Section 2.1.2.
Screening tests were also conducted by ARI/WSU to find a suitable site for
demonstrating that the tracer technique could be used to determine the
emissions from a complete distribution system. A requirement of the tracer
method is that the tracer and methane plume are uniformly mixed. This means
that the measurement must be made a sufficient distance downwind of the
source to allow the plumes to mix properly. The screening tests showed that
methane sources are so widely dispersed within a city that the methane plume
is only a few hundred ppb above the background level at the edge of the city.
The background concentration is typically 1850 ppb with a temporal
4-31
-------
variability of + 100 ppb. The plume from che city disperses rapidly and
cannot be distinguished from background levels long before the tracer gas
plume can be mixed with the methane plume from the city.
As described in References 8 and 9, a variety of modified tracer techniques
were formulated and explored. The basis o£ these methods is that an equation
can be derived from the Gaussian plume dispersion model which relaxes. the
requirement that the tracer and methane plumes be uniformly mixed. The
relationship for any point, line, or uniform area source is that the ratio
of emission rates (Em/Et) is equal to the ratio of the cross wind integral
of the concentration (ICM/ICT) at any point downwind of the source.
A test was conducted in a city which appeared to be best suited for
demonstrating the various modified tracer techniques based on screening tests
conducted in 15 cities around the country. This test indicated that whole
city methane emissions can be measured with roughly + 50 percent accuracy.
However, such measurements include all sources of methane, and it was
concluded that emissions from the natural gas system could not be determined
with sufficient accuracy for this program.
Emission Factor/Correlation Equations
A test program designed to develop new emission factors and a new correlation
equation was funded by GRI and the American Petroleum Institute (API). Star
Environmental is the contractor on the project.
Bagging and screening tests were conducted at 16 onshore sites in the four
major gas/oil production regions in the continental U.S. The 16 sites were
evenly divided among heavy and light oil, gas production, and gas processing.
Nearly 175,000 components were screened, and the emission rates were measured
for 750 randomly selected fittings using the bagging technique.
Bagging and screening tests were conducted on four offshore gas/oil
production platforms in the Gulf of Mexico. This program was designed to
complement the 1990 study sponsored by Mineral Management Services to collect
data • from gas/oil production platforms off the California coast
{Reference 10) .
Data on emissions and the percent of leaking components determined in this
program are compared in Table 2 with similar data collected by Rockwell for
API in the late 1970's. These data indicate that facilities are
significantly tighter now. About one third as many components leak, and the
emissions are a factor of 3 to 8 lower than the API/Rockwell data.
A plot of the correlation equations developed from the recent API/GRI data,
the API/Rockwell data, and the curve recommended by EPA are compared in
Figure 4. Because the correlation equation should be independent of the
leakiness of a facility, the correlation equations would be expected to be
the same. There is good agreement between the API/GRI and the older
API/Rockwell curves. However, the EPA curve would overpredict emissions from
the facilities tested by a factor of 3 to 10.
-------
Mass Balance Approach
The emissions from a transmission system are being calculated by Southwest
Research Inc. using the mass balance technique. The results from this method
will be available in October 1992.
Leak Statistics Method
This technique is costly to apply on a national scale. Therefore, a
cooperative study was formed with nine U.S., two Canadian, and two European
distribution companies to measure leakage from distribution systems. This
Leakage Measurement Program is intended to be a separate project designed to
provide participating companies a leakage estimate of sufficient accuracy
that operational issues can be addressed.
Data currently available are presented in Table 3 in aggregate and separately
for mains and services. As more data become available, the data will be
analyzed to determine the effect of age, material, pressure, size, and soil
characteristics as well as the leak survey technique used by the
participating companies.
Summary of Test Results
Tests conducted to determine the steady emissions for the various source
categories are summarized in Table 4. The table presents the number of
tests, the emission factor determined from the test data, the activity
factor, and a current estimate of steady emissions from the gas industry.
It is clear that much more data are required before a defensible estimate can
be made of emissions from U.S. gas operations. For example, the emissions
from only 13 meters in one distribution system have been measured out of an
estimated 50 million customer meters. The uncertainty in the estimate cannot
be evaluated. Although the random error in the emissions data is a
straightforward calculation, the bias cannot be addressed until data are
collected from other systems around the country.
2.2.2 Unsteady Emissions
The latest calculations of the unsteady emissions are summarized in Table 5.
As discussed previously, the unsteady emissions are a function of the cause
of the emissions from each source. Although as many as six calculations
would be required to account for all the unsteady emissions from each source,
most are negligible. As shown in Table 5, 97 percent of all unsteady
emissions are caused by normal operations, combustion, and maintenance
operations. Nearly 80 percent result from three activities, pipe and
facility blowdowns, compressor exhausts, and venting from pneumatic devices.
The calculated unsteady emission rates for a source are analogous to
measurements of the steady emissions. For the most part, information needed
to calculate unsteady emissions is obtained during site visits in which
information is also collected to quantify the activity factor. The columns
labeled "Number of Calculations" and "Number of Sites" in Table 5 indicate
the number of sources for which the emission rate was calculated and the
number of sites represented in the estimate.
4-33
-------
In developing the Phase III implementation plan, the variability in the
emissions is used to calculate the additional data needed to achieve the
accuracy target for each source. An analysis of the emissions data is also
used to assess bias and the need to collect data from a larger number of
sites. The results of this analysis are summarized in Table 5 under the
column, "Number Of Sites To Be Visited."
2.2.3 Total Emission Estimate
The unsteady emissions are estimated to be 102 BCF, and the estimate for the
steady emissions is 84 BCF. This gives a total of 186 BCF for the U.S. gas
industry or approximately 1 percent of gas production.
The uncertainty, the target accuracy of each source, and the number of
additional data points required to achieve the target accuracy are routinely
calculated each time a new emissions estimate is made. The current
uncertainty is around + 80 percent. However, this calculated uncertainty
includes only random error, not bias. The possibility of large bias errors
exists in the current estimate because of the limited number of sites
visited. For this reason, a defendable estimate of the uncertainty cannot
be provided at this time.
2 ¦ 3 Future Activities/Phase III
The purpose of Phase III is to collect sufficient emissions data to determine
emissions for the U.S. gas industry within 0.5 percent of production. The
flow chart for Phase III is presented in Figure 5.
The first task is to apportion available resources to focus on sources with
the largest emissions and highest uncertainty in the emissions estimate.
Each source is examined to determine if the accuracy target has been
achieved. If not, the number of additional data points needed to achieve the
accuracy target is calculated from the standard deviation for a 90 percent
confidence level. If there is more than one method of obtaining emissions
data, a cost optimization program is used to select the most cost effective
approach.
The most cost effective approach typically depends on the accuracy of the
method, the cost per data point, and the number of data points needed.
Because the standard deviation of the emissions data will change as
additional data are collected, optimizing the Phase III effort will be an
iterative procedure.
The second task of Phase III is to develop an implementation plan. The Phase
III implementation plan which defines the data needed and how, where, and
when the data will be collected, has been completed and documented (Reference
11) .
Testing started in the early summer of 1992 and will continue through the
summer of 1993. Facilities around the country will be visited to collect the
information needed to calculate the unsteady emissions and to quantify the
4-34
-------
activity factors. The unsteady emissions will be calculated and the steady
emissions will be measured at a sufficient number of sources to achieve the
accuracy target.
The last step of Phase III is to extrapolate the emissions data using the
emission and activity factors to obtain the national emissions estimate.
The current estimate of the additional measurements that are required to
achieve the accuracy target for each source is presented in Table 6. The
table also indicates which measurement techniques were determined to be the
most cost effective. In some cases, however, data will be collected
using more than one method in order to address the question of bias or to
improve the credibility of the data by demonstrating that different methods
give the same results.
Table 7 is a summary of the number of site visits that will be made during
Phase III and the information that will be collected to determine the
activity factors and to calculate the emission factors for the unsteady
emissions.
2 .4 Mitigation of Emissions
Emission estimates from the GRI/EPA program, although preliminary, were used
as the basis for a study conducted by Radian to examine cost effective
techniques for reducing emissions from the gas industry. This work was
sponsored by EPA and was not part of the GRI/EPA emissions study. A
cost/benefit analysis was conducted in which the "net present value" (NPV)
was calculated for 22 mitigation techniques. The net value was calculated
as the difference between the cost of implementing the control strategy and
the value of the gas saved. By ignoring any environmental benefit, the
analysis would indicate which control measures would make sense to implement,
based on reducing operating cost independent of the environmental benefit.
The assumptions used in the NPV analysis are:
° xhe value of benefits is only the value of the gas saved.
0 The average wage is $2 5 per hour and increases by 2 percent per year
in real terms (i.e., no inflation).
0 The discount rate and opportunity cost of equity is 6 percent in real
terms.
° The value of natural gas saved per 1000 cf is $1.59 for production,
$2.01 for transmission, and $2.91 for distributors. The value is
assumed to increase 2 percent per year in real terms.
The results of the study are summarized in Table 8 for the 22 techniques
evaluated.
The study indicates that improved inspection and maintenance programs,
4-35
-------
installation of low-bleed pneumatic devices, recompression of gas during
maintenance activities, pipe replacement, and use- of glycol dehydrator
exhaust as fuel could be cost effective in some circumstances.
The use of low-bleed control devices is found to be cost effective if
installed as the high-bleed devices are retired after a full lifetime. There
is only a small incremental capital and installation cost. The cost
effectiveness of inspection and maintenance programs is very sensitive to
labor cost and the amount of gas leaked. The use of glycol dehydrator
exhaust as fuel is predicated on having accumulators installed to capture the
hydrocarbons as part of a control program for toxic emissions that may be
required by new regulations. Pipe replacement is cost effective only in
areas with low installation cost ($80/ft) and a high repair rate (25 leak
repairs/mile). The reduction in gas lost had a negligible effect on, the
analysis of the pipe replacement strategy.
The cost effectiveness of recompression and reinjection of gas back into the
pipeline during maintenance operations varies from case to case. It is
generally sensitive to whether the compressor is rented or purchased and
whether the study assumes that the strategy will be implemented all year or
only when demand is low. Typically, a downtime of 4 days could be required
to recompress the gas, while venting the gas could take only 4 hours. The
difference in downtime during peak periods could be unacceptable.
Gasunie, a gas transmission company in The Netherlands, also conducted a
study to determine if recompression of gas was a cost effective approach of
reducing emissions (Reference 12). Because the compressor was rented, the
cost effectiveness of recompression depends on the amount of gas in the line.
In The Netherlands, recompression was found to be cost effective if more than
2.3 million cf of gas is recovered.
Since the benefit in the cost/benefit analysis is the value of the natural
gas saved, the results are obviously highly dependent on the accuracy of the
emissions data. This is a problem not only for the current study, but one
that is likely to affect future studies unless it is specifically addressed.
At this stage of the GRI/EPA emissions study, it is not possible to give a
credible estimate of the uncertainty in the emissions data. This will be
possible as the study progresses further into Phase III. However, this
program was never designed to determine emissions from specific types of
sources with a high degree of accuracy.
The goal of the program is to determine overall emissions to within 0.5
percent of production. This can be achieved even if the error bound on the
emissions estimate for specific source categories is + a factor of 2. This
is not sufficiently accurate for a cost/benefit analysis to decide which
mitigation strategies are cost effective.
To provide sufficiently accurate data for these studies would be cost
prohibitive. There seem to be two alternatives. The first is to improve the
accuracy of the emissions estimate for a targeted set of sources that are
candidates for mitigation measures. The second is to reverse the roles and
use a cost/benefit analysis to determine the emissions that are required for
4-36
-------
specific mitigation strategies to be cost effective. These results would
provide valuable guidelines to the industry on ways of reducing operating
cost. Emission rates vary substantially from site to site. Although a
specific mitigation technique may not be cost effective for the average
facility, many companies could find it very profitable to implement the
technique because their emissions are much higher than the average value.
As the industry implements these cost saving techniques, their feedback will
provide the accurate, detailed information needed to refine the mitigation
strategies.
4-37
i
-------
SECTION 3.0
SUMMARY / CONCLUSIONS
The primary purpose of the study is to provide an emission inventory accurate
enough for global climate modeling studies and for addressing the policy
question of whether or not to encourage the increased use of natural gas in
order to reduce global warming.
The GRI/EPA program is designed to determine methane emissions from the U.S.
gas industry within +0.5 percent of gas production. The purpose of Phase
I was to define the problem. Phase II was spent developing the tools needed
to measure or calculate methane emissions from the various source types that
comprise the industry. Also, because of the large number of sources within
each source type, techniques were developed for extrapolating emissions data
to similar sources within the industry. The purpose of Phase III is to
collect sufficient data to achieve the accuracy goal of the study.
Emissions data are also needed for examining cost effective techniques for
mitigating emissions from the gas industry. Unfortunately, it would be cost
prohibitive to determine the emissions from all sources with the accuracy
needed for cost/benefit studies. This problem can be alleviated by improving
the emissions estimate for a targeted group of source types. However,
another approach would be to reverse the roles, and use the cost/benefit
analysis to determine the emissions that would be required for selected
mitigation strategies to be cost effective.
Phases I and II have been completed. Phase III was started in early 1992,
and the data collection effort will be completed in the summer of 1993. The
entire study is expected to be completed by early 1994.
Based on the limited amount of data collected to date, methane emissions for
the U.S. gas industry appear to be in the range of 1 percent of production.
The estimate of the unsteady emissions is 102 BCF, and the steady emissions
are estimated at 84 BCF per year.
4-38
-------
REFERENCES
1. U.S. Environmental Protection Agency, "Protocol for Generating
Unit-Specific Emission Estimates for Equipment Leaks of VOC and
VHAP, " EPA/450/3-88/010, NTIS PB89" 138689, 10/88.
2. Radian Corporation, "Screening and Bagging of Selected Fugitive
Sources at Natural Gas Production and Processing Facilities," Field
Test Report, June 1990.
3. Uthe, E.E., W. Viezee, and L.J. Salas, "Evaluation of Methane
Emissions from Natural Gas Production Operations Using Tracer
Methodologies," Final Report GRI-93/0102, March 1993.
4. McManus, J.B. et al. , "Measuring Urban Fluxes of Methane," World Res.
Rev. 3, pp. 162-183 (1991).
5. Pacific Gas and Electric Co., "Unaccounted-For Gas Project,"
GRI-90/0067.1, Five Volumes, June 1990.
6. McManus, J.B. et al., "Methane Emissions from Natural Gas Systems,"
Aerodyne Research Inc., Report No. ARI-RR-931, June 1992.
7. Lamb, B. et al., "Measurement of Methane Emissions from Natural Gas
Systems Using a Tracer Flux Approach," 85th Annual Meeting Air and
Waste Management Association, June 1992.
8. McManus, J.B. et al., "Urban Methane Measurements Field Campaign #1:
Midwest Cities," Aerodyne Research, Inc., Report No. ARI-RR-920,
April 1992.
9. McManus, J.B. et al., "Demonstration of Methane Emission Measurement
Techniques for an Urban Natural Gas Distribution System and a Natural
Gas Production Facility," Aerodyne Research, Inc., Report No.
ARI-RR-945, August 1992.
10. Countess, R.J., and D. Herkhof, "Fugitive Hydrocarbon Emissions from
Pacific OCS Facilities," 84th Annual Meeting Air and Waste Management
Association, June 1991.
11. Radian Corporation, "Phase III Program Plan - Implementation," Report
DCN 92-263-081-02, June 1992.
12. Veenstra, T. , et al. "Methane Reduction in the Gas Transportation
System," 85th Annual Meeting Air and Waste Management Association,
June 16,1992.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's
peer and administrative review policies and approved for presentation and publication.
^-39
-------
PRODUCTION
r
WELLS
j QATXERINq
j separation"
[
OAS PLANTS
] [
I OAS INDUSTRY TOT All
TRANSMISSION
c
PIPELINES
COMPRESSORS
]-
[
METERS
REGULATORS
] t
1 PIPING | | VALVES | IfLANQEsI | SEALS |
DISTRIBUTION
MAINS
SERVICES
REGULATORS
METERING
STATIONS
METERS
I
o
Figure 1 U.S. Gat Industry Ditaggregatad by So urea
SEGMENTS/
SOURCE TYPO
TECHNIQUES
EMISSION
FACTOR
CORRELATION
EQUATION
TRACER
OAS
UEAK STATISTICS
MASS BALANCE
pwuoucnuN
• WEILS
• '
•
* OATHEMNO
*
O
• SEPARATION
•
•
•
• OAS PLAWTS
•
•
•
TRANSMISSION
o
• PIPEUNES
*
• compressor!
•
•
• METERS
•
* REGULATORS
•
•
DISTRIBUTION
•
* MAMS
•
• SERVICES
•
• METERIWO
STATIONS
*
*
> REGULATOR
STATIONS
•
•
* METERS
•
4
O MMOOMOI
• COMTD
* BASED ON data moniwmwnoiiewEMi
Flpna TocMqunD*MtopMffarMaauhaSteadyEmissions
PHASE II
METHODS DEVELOPMENT
DEVELOP EXTRAPOLATION
TECHNIQUES
VALIDATION
PHASE HI
DEVELOP
METHODS FOR
CALCULATING
UNSTEADY
EMISSIONS
DEVELOP METHODS FOR
MEASURING STEADY EMISSIONS
CHARACTERIZE
GAS INDUSTRY
i Type 1 Number
of Bourns
• Operating Procedure*
• MaMonane* ActMtlei
FTgurs 2 Flow Chart of Ptuee II
TABLE I: MEASURED METHANE EMISSIONS FROM METERING
AND PRESSURE REGULATING STATIONS
NUMBER
EMISSION
ACTIVITY
EMISSIONS
TOTAL
CATEGORY
OP TESTS
PACTOR (CF/HR)
PACTOR
(BCP)
EMISSIONS
NO OF
(BCP)
FACILITIES
All
J9
120
135.000
140
140
Ref. Station!
27
65
132.000
75
MAR Suilom
12
260
2,600
6
II
Reg. Suihxn:
> 300
12
•7
3,700
2 8
100 lo 100
7
«5
11.000
13.6
40 to 100
4
0S0
90.000
0.6
< 40
4
0.90
20,000
02
MAR Sutkra:
> J00
a
370
14,000
4 5
100 lo 100
i
390
1,000
5
40 to 100
3
0 05
200
0.0
26.4
< 40
0
0
0
0
-------
TABLE 2: COMPARISON OP BAGGING AND SCREENING DATA
TYPE PACILITY
DATA SET
% COMP.
LEAKING
EMISSIONS
LBS/COMP.-DAY
Gas Plinti
API/Rockwell
12.0
0.130
API/OR! 1992
4.2
0.017
Gts Production
API/Rockwell
4.7
0070
API/GRI 1991
3.4
0.020
Off-Shore Pisiform
Oulf of Meilco
API/Rockwell
3.1
0.017
API/ORl 1992
I.I
0 007
TABLE 4: SUMMARY OF STEADY EMISSIONS
EMISSION
NUMBER OP
FACTOR
ACTIVITY
EMISSIONS
CATEGORY
TESTS
(CFOTR)
PACTOR (•)
(BCP)
Production/Processing:
Wells
321
2,000
540,000
1.0
Gathering Lines
• •
20,000
91,000
1.8
Separators
114
8.000
10,000
1.4
Healers
84
9,000
34.000
0.5
Dehydrstors
32
11,000
20.000
0.2
Meiers
13
20,000
20,000
0 4
Gas Plsnls
42
1.0 X l(f
742
0.7
Transmission:
Compressor Stations
0
6.3 X 10*
2,000
13.0
MAR Stations
8
3X 10*
5,000
16.0
Distribution:
Mains
37
20,000
136,000
16.3
Service*
30
63
43 X 10*
3.0
Pressure Reg > 100
19
800.000
22,000
160
Pressure Reg < 100
8
8,000
110,000
0.8
MAR > 100
1
5X I0»
1,000
3.0
MAR < 100
3
500
200
00
Customer Meters
13
140
50 X IC
70
TOTAL:
83.3
* Number of facililic* or miles of pipe for mains and gathering lines
•* Based on dm for mains
0.1
0.01
0.001
!
0.0001
100
1000
100000
10000
Instrument Screening Value (ppmv)
EPA Flanges API/Rockwell APt/GRI
Figure 4 Correlation of Instrument Screening
Values to Total Hydrocarbon Emissions
TABLE 3: LEAKAGE MEASUREMENTS OF BURIED
PIPE IN DISTRIBUTION SYSTEMS
CATEGORY
NUMBER
OF TESTS
AVERAGE
(CF/HR)
COEFFICIENT
OF VARIATION
ALL
62
2.3
3.4
MAINS
30
3.6
3.2
SERVICES
32
1.2
1.4
-------
TABLE 5: CALCULATED VALUES OF THE UNSTEADY
EMISSIONS FROM U.S. GAS OPERATIONS
NUMBER
OF SITES
* OF
% OF
NUMBER OF
NUMBER
TO BE
EMISSIONS
TOTAL
UNSTEADY
CATEGORY
CALCULATIONS
OF SITES
VISAED
(BCF)
EMISSIONS
EMISSIONS
Normal Operations:
Pneumatics
26
26
31
37
19
36
Compressor Start/Stop
1
1
37
8
4
8
Dehydrator Vents
2
2
23
9
5
9
Combustion:
Compressor Exhausts
861
112
37
22
12
22
Maintenance Operations:
Well Workovers
2
2
23
2
1
2
Facility & Pipe Blowdowns
1
1
27
21
11
20
Mishaps:
¦Dig Ins*
1
1
27
3
1.5
3
Upsets
—
—
56
-
—
— .
Leaks
—
—
9
-
—
-
TOTAL:
56*
102
54
100
• The total consists of visits to 9 distribution systems, 18 transmission segments, 23 production fields,
and 6 processing plants.
PHASE III
IMPLEMENTATION
EXTRAPOLATE
EMISSIONS
COLLECT
INFORMATION
CALCULATE
UNSTEADY
EMISSIONS
MEASURE
STEADY
EMISSIONS
DEVELOP
RATIONALE
FOR
APPORTIONING
RESOURCES
DEVELOP
PHASE III
IMPLEMENTATION
PLAN
Figure 5 Flow Chart for Phase III
4-42
-------
TABLE 6: SUMMARY OF PHASE m TEST PLANS
SEGMEXTS'
SOURCE TYPES
BAGGING
EMISSION
FACTOR
CORRELATION
equation
TRACER
LEAK
STATISTICS
Production
- Wells
321 0
4 0
0 4600
0 23
- Separators
114 0
4 0
0 920
0 23
0 18
0 6
- Heaters
84 0
4 0
0 230
0 23
0 12
0 6
- Dehydrators
32 0
4 0
0 184
0 23
0 6
0 6
- Meters
15 0
2 0
0 1380
0 23
0 12
0 6
- Gas Plants
4 0
4 0
0 6
0 6
1 3
1 3
Transmission
- Compressor
Stations
0 20
0 6
-M&R
Stations
7 24
6 6
Distribution
- Mains
30 100
2 13
- Services
32 100
2 13
- Metering
Stations
5 30
3 6
- Regulator
Stations
27 45
11 6
- Meters
26 0
2 0
100 1000
2 5
1 - Ho.
2 - No.
3 - No.
4 « No.
Of Tests
of Sites
To Be Tested
of Sites
-------
TABLE 7: SUMMARY OF SITE VISITS AND INFORMATION TO BE GATHERED
StTBTYPB
# OP StTBS
visrrED
§ OF SITES
TO BE
VfSTTED
Activity factor data
EMISSTOti PACTOR DATA
Gas/Oil Field
2
23
Separator* per well
Healers per well
Field throughput
Dehydrators per well
Vessel blowdown frequency
Compressor start/stop frequency
PRV* lift frequency
Pneumatics per equipment type
Number of compressors, type
Dehydrator conditions
Vessel volumes
Compressor volumes
Pield pressure
Pneumatic device types
Compressor exhaust, fuel use
Gathering
Facilities
0
20
Station throughput A mileage
Line blowdown frequency
Compressor start/stop frequency
Pneumatics per equipment type
Number of compressors, type
Volume of lines
Compressor volumes
Blowdown volumes
Gas Plants
3
6
Maintenance frequency
Compressor start/stop frequency
Number of compressors, type
Dehydrator count
PRV lift frequency
Plant volumes A practices
Compressor volumes
Compressor exhaust, foe! use
Dehydrator conditions
Transmission
Facilities
1
18
Throughput and mileage
Compressor start/stop frequency
Number of compressors, type
Pneumatics per equipment type
PRV lift frequency
Mishap frequency
Compressor volumes
Compressor exhaust, ftiel use
Pneumatic device types
Mishap volumes
Distribution
Systems
2
9
Mileage main & services, type
Number class 1 leaks
PRV lift frequency
Mishap frequency
Pneumatics per equipment type
Blowdown volumes
Pneumatic device types
Mishap volumes
Pressure TTcJTcT^alve1
-------
TABLE 8
SUMMARY COMPARISON OF ECONOMIC COSTS AND BENEFITS
FOR METHANE EMISSION REDUCTION TECHNIQUES
D*
Shaft
No.
Tednuqut
K*Prt*»t
Vnk*
• (MiTFvmi S)
lAcrtmcafcri
JwBCF
ftrarfoetft)
Go* per
fisxrvity
Unit {$)
Adiviry
; . Aooutl
: RedwaioB : :
{BQPtyr)
jfcODucnon
i
rortinr> «ad bbmuoi of tntag tmyuunu-
4-18
+360
+4^49
23
Suiion
Mik
11.960
IS
CMfal rf —h— je—m fay tMhwinainri rf hm'-fakad pnwk imAmt
(or ni<
-1,891
Debydfttor
1 172
13-
UM| ate ciJawi fr» fiyai Mydnun m boikr hm\ amwiiktan
bpfcoc
+6.4
+27
+ 76
Dct^drtUM
IJ72
16"
«.n»n»iUiw iddiliflBd
-160
-690
-1.919
Odjydntn
1.872
ir
Use of pnruhfc oanptttMon lo f (*vr warned f*s during inmcvi—k»
tiwUeedew
+0.743
+ 10
+72
+0 J*8
Bloedown
Mik
9360
i«*
Control of oadbBv ban dure* Ike kkmAavu of pipe lino* for
nuke nkkaaBK) ikau|b lbs iac of mMIc Ikm
¦6J
-30
-134
-7,118
BMowd
Fbuc Unit
11.440
19
famfiag • fM ttnfcinc to • Hrtpnvihn oogiae m a ne*
+0.173
N/A
+12-52
l^-yr
0.090
20
CobjqI of oaftHwr «Mim ifarau^ wurjfk «mh oualyiie uimiiui
-450
-104*
-9J7
14.903
CaqnMor
3.252
UtSTRIBUnOW
21
FMfHfHnr -diiiT^ taBpeoiOB and anamac—arr of kaknif osnptmu
+x
+ 320
+ 1^380
Sution
4^12
22
High:
+* 1,400
¦I 43.000
+ 1«,OCD.OOO
-5,158,000,000
+S 1,400
4 43,000
Mik
Mik
8 J2E-6
"Ciyvol dcfcydftiov apuai ut vutil!)1 a«it0h« withm oocfe wior
"Mobtk faim nd oaapmuMf* ra> be u«d v uuvkro or acpinticly, but loul
4-*5
£
i
£
i
be more \tmo 11.44 BCF/yr.
-------
EMISSIONS AND MITIGATION AT LANDFILLS
AND OTHER WASTE MANAGEMENT FACILITIES
Susan A. Thorneloe
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
4-46
-------
ABSTRACT
Landfills and other waste management sources of methane are amenable to cost-effective
control. Consequently these sources have been give a high priority for clarification of their
emission potential. "Hie United States Environmental Protection Agency (U.S. EPA) is conducting
research to determine the emission potential and mitigation opportunities for cost-effective control
for the major sources of greenhouse gases. EPA's Air and Energy Engineering Research
Laboratory (AEERL) is responsible for developing more reliable global and country-specific
estimates for the major sources of greenhouse gases including waste management, coal mines,
natural gas production/ distribution, energy usage, cookstoves, and biomass combustion. AEERL
has gathered data which have resulted in the development of more reliable estimates for landfills.
Research has been initiated to characterize the methane potential of other waste management
facilities including wastewater treatment lagoons, septic sewage systems, and livestock waste.
AEERL is also documenting the current state of technology for utilization projects.
Currently there are 114 landfill gas to energy projects in the U.S. and about 200 worldwide.
Technology transfer/technical assistance programs have been initiated to help encourage the
utilization of waste methane and to help implement the upcoming Clean Air Act (CAA) regulations
for municipal solid waste landfills. For example, AEERL is working with a consortium of local
government representatives to explore the application of EPA research on methane/energy recovery
from municipal solid waste landfills. AEERL also serves on the International Energy Agency
Expert Working Group on Landfill Gas and the Steering Committee for the Solid Waste
Association of North America. AEERL is also responsible for demonstrating innovative
approaches to the control of waste methane such as the application of fuel cell technology to
recover energy from landfill gas and digester gas.
This paper describes the emission potential for waste management sources and the
mitigation opportunities. It also provides an overview of some of the barriers in the U.S. that
affect methane utilization. This research is funded through EPA's Global Climate Change
Research Program. This paper has been reviewed in accordance with EPA's peer and
administrative review policies and approved for presentation and publication.
4-47
-------
INTRODUCTION
Waste disposal results in emissions of greenhouse gases including methane (CH4), carbon
dioxide (CO2), nitrous oxide (N2O), ozone precursors, and chlorofluorocarbons. The major
sources of CH4 from waste management include landfills, wastewater treatment lagoons, and
livestock waste. Current estimates suggest that this source accounts for up to 125 Tg/yr or -40%
of the estimated total global anthropogenic emissions of 300 Tg/yr (IPCC, 1992). Landfills have
been estimated to contribute as much as 60 Tg/yr of CH4. Policies are being considered to reduce
greenhouse gas emissions to meet the goals of die United Nations Conference on Environment and
Development held in Rio de Janeiro in June. Emissions sources that are amenable to control -
such as landfills - have been given a high priority for clarification. (EPA, 1989)
"Waste" CH4 results from the anaerobic decomposition of biodegradable waste found in
landfills, open dumps, waste piles, wastewater treatment lagoons, septic sewage systems, and
livestock waste. This waste CH4 can be a source of pollution as well as a resource. There are 114
landfill gas (LFG)-to-energy projects in the U.S. (Thomeloc, 3/92) and 200 LFG-to-energy
projects worldwide (Richards, 1989). Landfill gas is utilized (1) as medium-heating-value fuel,
(2) to generate electricity using internal combustion engines, or gas and steam-fed turbines, and (3)
as high-heating-value fuel in which case the gas is upgraded and fed into a nearby natural gas
pipeline. U.S. landfills cunrndy generate 344 MW of electricity (Thomeloe, 3/92). The gas that
is formed from anaerobic decomposition is typically 50 to 55% CH4,45 to 50% CO2, and <1%
trace constituents.
The CH4 is a concern because of its global wanning effects and explosive potential.
Emissions of nonmethane organic compounds (NMOC) contribute to tropospheric ozone which
aggravates urban smog and is a concern to human health and the environment. Other LFG
constituents such as vinyl chloride, benzene, carbon tetrachloride, and methylene chloride are a
concern for their cancerous and noncancerous effects. The Agency has proposed CAA regulations
for emissions from municipal solid waste (MSW) landfills (FR, 1991) which will reduce five
health and welfare effects: (1) explosion hazards, (2) global warming effects from CH4 emissions,
(3) human health and vegetation effects caused by ozone fonned from NMOCs, (4) carcinogenicity
and other possible noncancerous health effects associated with specific landfill emissions
constituents, and (5) odor nuisance (U.S.EPA, 3/91). Estimates from the proposed regulations
indicate that 621 landfills of the 6,000 existing active landfills would be required to collect and
control MSW landfill emissions (p. 24480, Hi, 1991).
The proposed Clean Air Act regulations do not require utilization of the gas. Although
increased CO2 emissions are being traded off for reduced CH4 emissions, there is a net benefit due
to the difference in the radiative forcing capacity between CO2 and CH4. The radiative forcing
capacity of CH4 to CO2 on a molecular basis is 21 times that of CO2 (p. 53, IPCC, 1990). It is
hoped that the sites affected by these regulations will consider LFG to energy as opposed to flaring
the gas. The use of energy recovery for the control of MSW landfill air emissions will result in
decreased emissions of CH4, NMOCs, and toxics. Additional benefits include the conservation of
global fossil fuel resources, reduction of emissions at coal-fired power plants, reduced dependency
on imponed oil, and cost savings to public entities that receive royalty payments (Thomeloe, 6/92).
However, there are many barriers in the U.S. associated with the utilization of waste CHi.
This paper provides data and information that characterize the CH4 potential of different
waste management sources including landfills, wastewater treatment lagoons, septic sewage
systems, and livestock waste. Mitigation opportunities are identified and the different options are
described. Bamerc that affect waste CH* utilization are identified. This paper describes research
that is being conducted through EPA's Office of Environmental Processes and Effects Research on
4-48
. Ik
-------
Global Climate Change. AEERL has responsibility for characterizing the emission potential and
investigating mitigation opportunities for the major sources of greenhouse gas emissions
(Thorneloe, 1991).
"Waste" CEL Potential from "Waste" Management Sources
The major sources of CH4 from the anaerobic decomposition of waste include landfills,
wastewater treatment lagoons, septic sewage systems, and livestock waste. Table 1 presents an
estimate of the relative contributions of each of these sources for the U.S. and globally. These
estimates suggest that these sources on average account for 80 Tg/yr or -30% of the total global
anthropogenic emissions of 300 Tg/yr (IPCC, 1992).
Table 1. U.S. and Global Estimates (Tg/yr) of "Waste" Methane Emissions
U.S. Reference Global Reference
Ave. Ranee Avg, RftPgC
Landfills 9 (6-13)* U.S. EPA, 7/92 30 20-70 IPCC, 1992
Wastewater Treatment/
Sewage Treatment ? 25 ? IPCC, 1992
Uvestock Waste 4 Safle, 1992 25 (20-30) IPCC, 1992
•Potential emissions, not corrected for the amount that is flared or utilized. Approximately 1.2
million tonnes of CH4 is being recovered from U.S. landfills (Thomeloe, 3/92).
Estimates of global CH4 emissions were summarized by the IPCC and suggest that
landfills contribute ~30 Tg/yr with a range from 20 to 70 Tg/yr (p. 35, IPCC, 1992, Khalil and
Rasmussen, 1990). Preliminary estimates generated using AEERL's empirical model indicate that
potential landfill CH4 emissions in the U.S. range from 6.3 to 13 Tg/yr, with an average of 10
Tg/yr. Global estimates suggest a range of 20 to 40 Tg/yr of CH4 emissions with an average of 30
Tg/yr. Estimates generated using Bingemer and Crutzen's approach — which is currently proposed
as the official IPCC methodology (OECD, 1991) - indicate that landfill CH4 emissions contribute
60 Tg/yr globally and 23 Tg/yr in the U.S. (U.S. EPA, 7/92).
The estimates generated using the empirical model are thought to more accurately reflect the
amount of CH4 from landfills that is contributing to the global CH4 flux (Campbell et al., 1991,
Peer et al., 3/92, Peer et al., 1992). The estimate using the empirical model uses data from landfill
gas recovery systems and accounts for CH4 oxidation and gas recovery efficiency. The data that
were used to develop the empirical model were collected from over 100 U.S. landfills. An EPA
report is being published that documents the development of the model and the estimate of CH4
emissions for U.S. landfills. Future refinements of this estimate will adjust for waste composition
using data being developed on the gas potential of different biodegradable waste streams.
Estimates for wastewater treatment are less reliable primarily due to a lack of country-
specific data needed to characterize the CH4 potential of municipal and industrial wastewater
treatment There is also a lack of field data characterizing the CH4 potential from lagoons
(Thomeloe, 2/92). Lagoons (or surface impoundments) are usually earthen pits used to contain
and process wastewater. AEERL is initiating a field test program in 1993 to collect lagoon
characterization data such as the biological oxygen demand (BOD) loading, flow rates, and
4-49
-------
retention time. These data will be used to develop a methodology for estimating greenhouse gas
emissions from lagoons including CH4, CO2, N20, and NMOC emissions. Initial estimates far
this source suggest that CH4 emissions range from 10 to 40 Tg/yr (EPA, 7/92).
CH4 emissions from wastewater treatment lagoons are not expected to be a major source in
the U.S. since many digesters flare and sometimes utilize the gas to control hydrogen sulfide
emissions. However, lagoons may be a more significant source in developing countries where
lagoons are being more frequently used and the gas is not controlled. Agencies such as the World
Bank (Bartone, 1990) recommend die use of lagoons for wastewater treatment for developing
countries since land space is readily available, operation is relatively simple, cost is low, and
energy requirements are minimal. This represents a potential opportunity to work with developing
countries to demonstrate that the CH4 can be utilized as an alternative energy source.
Individual onsite wastewater treatment systems, such as septic systems, are used
throughout the world. In China, there are an estimated 10 million biogas pits, which are designed
to produce biogas for household use. However, the majority of the world does not collect the gas
from septic systems. A portion of this CH4 will be oxidized and some will be emitted to the
atmosphere. Field test work by F.PA/AFFRI. is planned in FY94 to collect data that will result in
more reliable estimates for this source and to determine if this source is amenable to cost-effective
control.
The only published global estimate for livestock waste suggests that CH4 emissions from
this source are about 28 Tg/yr with a range of about 20 to 35 Tg/yr (Safle et al., 1992). These
estimates were made by collecting information from animal waste management systems and the
quantity of animal waste managed by each system. Information was also collected from
government statistics and literature reviews. The major uncertainty regarding these estimates is due
to the assumptions and data characterizing the CH4 potential from the waste of free-range animals.
AEERL is planning to conduct laboratory and field studies in FY93 that will help reduce the current
uncertainty with these estimates and will investigate opportunities for cost-effective control.
MITIGATION OPPORTUNITIES FOR "WASTE" METHANE
The recently proposed regulations for MSW landfills will result in the reduction of 5 to 7
millions tonnes of CH4. Currently U.S. landfills are recovering 1.2 million tonnes of CHt and
producing 344 MWC of power. In the U.S., there are 114 LFG-to-energy projects (Thomeloe,
3/92). The breakdown of these projects by energy utilization option is presented in Figure 1. The
majority of these projects (ie., 75%) generate electricity which is either used onsite or sold to a
local utility. These projects are located across the U.S.in 28 states, with 38 LFG-to-energy
projects in California and 14 in New York. Of the 24 projects using turbines, 21 projects are gas-
fed and 3 projects are steam-fed. The largest project in the world is the Puente Hills Landfill in
Whittier, California. It is operated by the Los Angeles County Sanitation Districts and generates
50 MW, (Valenti, 1992). The majority of projects that produce electricity (i.e., -80% or 66 out
of 85) produce 1 to 5 MWe (Thomeloe, 6/92). Typically there are three to five engines or one to
two turbines per project.
4-50
-------
Figure 1. Number of U.S. Landfill Gas Projects by Energy Utilization Option
(Source: Thomeloe, 3/.92)
3
>.
81
©
& High Heating-Value Gas
UJ
o
®
.S
J> Med Heating-Value Gas
c
c
IC Engines
Turbines
61
0 10 20 30 40 50 60 70
Number of LFG Projects
Globally there are about 200 LFG-to-energy projects (Richards, 1989). There are also
digesters in use for livestock waste and wastewater treatment sludge that utilize the waste CH4 for
producing steam or electricity. Data are not presendy available to calculate the extent of digester
gas utilization. The Expert Working Group on Landfill Gas of the International Energy Agency is
compiling data on waste CH4 projects (Lawson, 1992). These data will be available in the future
to adjust current estimates of "waste" CH4 for the amount that is controlled. In the U.S. 12
millions tonnes of waste CH4 is being utilized by LFG-to-energy projects (Thomeloe, 3/92).
The utilization of waste CH4 can also result in a substantial cost savings to public entities
that own landfills and receive royalty payments. For example, Pacific Energy — who has
developed 25 LFG-to-energy projects - has paid out $ 13 million in royalties, mostly to public
entities. On average, Pacific Energy's projects are in the sixth year of operation under anticipated
20-year project lives (Wong, 1992). Other economic benefits include the purchase of goods and
services. In 1991, Pacific Energy purchased over $4 million in outside goods and services to
support its LFG projects plus a payroll of >$3 million. LFG to energy projects tend to be capital
intensive and are typically built on what is considered undevelopable acreage. Pacific Energy's
eight LFG-to-energy projects in California pay >$350,(XX) per year in property taxes in California
and require few public services (Wong, 1992).
There are emerging technologies for waste CH4 utilization. For example, AEERL initiated
a project in 1991 to demonstrate the use of fuel cells to recover energy from landfill gas. There are
a number of advantages with the use of fuel cells including higher energy efficiency, availability to
smaller as well as larger landfills, minimal byproduct emissions, minimal labor and maintenance,
and minimal noise impact (i.e., because there are no moving parts). The type of fuel cell being
demonstrated for LFG application is the commercially available 200 kWe phosphoric acid fiiel cell
(PAFC) power plant The 1-year full-scale demonstration is scheduled for 1993 (Sandelli, 1992).
4-51
-------
The major technical issue associated with the application of fuel cell to LFG is finding a gas
cleanup system that effectively and economically cleans the gas to the fuel cell's stringent
requirements. LFG composition can be quite variable as to the type of constituents and
concentration. Chloride and sulfur compounds are quite common. "Slugs" of condensate have
also been known to cause havoc at gas turbine and internal combustion engine projects (Augenstein
and Pacey, 1992). If this project is successful, it will provide a more environmentally attractive
option for waste CH4 utilization that is also more energy efficient.
A second project has been initiated by AF.ERL to demonstrate the use of fuel cells to
recover energy from digester gas. Digesters are frequently used at wastewater treatment (WWT)
facilities to process sludge. "Die digesters can utilize the waste heat from the fuel cells which
would result in an energy efficiency approaching 80%. The same type of commercially available
fuel cell will be used. The three issues to be addressed in this project are (1) feasibility of
integrating the PAFC power plant operation with the WWT plant, (2) anaerobic digester gas/waste
CH4 cleanup and processing requirements for fuel cell operation, and (3) improved fuel cell
performance on reduced heating value fuel (i.e., waste CH4 versus natural gas). This project is to
begin this fall and a 1-year demonstration is planned for FY94. It is expected that the LFG cleanup
process can also be used for digester gas. Generally digester gas trace constituents are less
variable than LFG although concenirations of sulfur compounds tend to be greater. The first phase
of the project will evaluate the requirements for the digester gas cleanup process and the application
of fuel cells to recover energy from a sludge digester at a WWT plant The design specifications for
the digester gas/fuel cell system will be provided in the Phase 1 Report.
Other emerging technologies for landfill gas include the production of liquid diesel fuel
such as the process in Pueblo, Colorado, that began operation last year, A second site in the U.S.
has been proposed to produce vehicular fuel from landfill gas. The South Coast Air Quality
Management District has awarded a contract to demonstrate a process for producing methanol from
landfill gas. The site selected for this demonstration is the BKK landfill, where there was co-
disposal of hazardous and municipal waste. TeraMeth Industries is responsible far the
demonstration which is scheduled to begin in 1993.
To help promote and encourage landfill gas utilization, case studies of six sites were
conducted in FY91/92. The final report (Augenstein and Pacey, 1992) contains detailed
information on the six LFG-to-energy projects. In addition, the report provides information on a
project recently developed by Michigan Cogeneration Systems (Appendix N), 25 Waste
Management LFG-to-energy projects (Appendix M), 11 Laidlaw LFG-to-energy projects
(Appendix L), and 4 case studies of United Kingdom LFG-to-energy projects (Appendix K). This
report is being referenced in the upcoming CAA regulations for U.S. MSW landfills as an
"enabling" tool that provides up-to-date information on LFG utilization for landfill owners and
operators. The report has generated a great deal of interest both in the U.S. and internationally
through the International Energy Agency, The International Solid Waste Association, and the Solid
Waste Association of North America.
A follow-up technology transfer project is focusing on the technical issues associated with
LFG cleanup and energy equipment modifications (i.e., application to LFG versus natural gas).
There are different philosophies associated with "waste" CH4 utilization. Information provided by
industry experts in the U.S. and Europe is being collected. These projects have been in existence
since the early 1980s and much has been learned as to what constitutes a "successful" project. The
EPA report that is scheduled to be published in the Fall of 1993 will review the current state of
knowledge for successful waste CH4 utilization projects. This technology-transfer project is
intended to help ensure that future utilization projects are designed and operated using the must up-
to-date knowledge and information on gas cleanup and energy equipment modifications.
4-52
-------
BARRIERS TO "WASTE" CH4 UTILIZATION IN THE U.S.
A major factor in helping to encourage LFG-to-energy projects is the Public Utility
Regulatory Policy Act (PURPA). It guarantees that utilities will purchase power that was
generated from landfills at a price related to the costs that utilities would experience to produce the
same amount of power. Although this guarantees a purchaser for the power, the power sale
revenues may be low if the utilities' own generating costs are low. This is the case in pans of the
U.S. where electricity is generated using water power (i.e., hydroelectric). However, with current
low energy prices, most regions of the U.S. also have unattractive "buy-back rates" for the
electricity that is generated using waste CH4. In addition, incentives such as federal tax credits that
have helped to encourage these projects appear to being losing favor. Industry experts think that
many of the marginal projects cannot continue without these tax credits. Current trends are toward
lower energy prices, reduced tax incentives, and increasing environmental liability.
Although there are more than 6000 landfills in the U.S., there are less than 120 LFG-to
energy projects. During the oil crisis in the 1970s/1980s, when the price of oil increased from $6-
8 per barrel to $35 per barrel, there was much more interest in developing alternative sources of
energy such as LFG-to-energy projects. With the current prices of energy, it is much more
difficult to find projects that are economical. Most U.S. projects that have had to cease operation
did so primarily due to economics. Projects that are upgrading the gas to pipeline quality have
been especially hard hit due to high operating costs and low revenue. Projects of that type are not
being planned in the U.S. However, sites in the Netherlands are finding more favorable
economics (Scheepers, 1991).
Laidlaw Technology Inc. suggests that "successful" LFG projects need to be over 1 MWe
and have an electrical price of at least $0.06-0.07/kWh including any capacity payments. Royalties
should not exceed 12.5% at this energy pricing (Jansen, 1992). Laidlaw also suggests that, if
higher royalties are offered, the percentage should be a function of energy pricing over and above
the base energy rate as inflation occurs. The early LFG projects were based on an established firm
price for net energy which provided a substantial degree of security to developers. Contracts for
many LFG projects do not allow for fluctuations in energy rates and costs. Revenues for energy
sales are usually based on prices of the "competition" of equivalent energy sources (e.g.,
petroleum products). Since the value of the energy base commodity can fluctuate, this can impact
profit.
Administrative and development costs have increased as revenues have decreased. These
costs include legal fees, permit applications, and contract negotiations for gas lease agreements and
power purchase agreements. These costs may vary widely depending on the environmental issues,
development considerations, and regulatory requirements. John Pacey of Emcon Associates has
found that these costs can vary from $30,000 to $1,000,000 per kWh for a 1 MWe LFG-to-energy
project (Augenstein and Pacey, 1992). The costs for a 1 MWe project are summarized in Table 2.
The gas extraction/collection systems are less than 15% of the total cost of the project. The major
cost component is the electricity generating equipment.
Tax credits are proportional to gas energy delivery as legislated by Congress (Section 29 of
the IRS Code) in 1979 to encourage non-fossil fuel use. These credits are a direct offset to taxes
and can be used only to offset a profit. The tax credits will extend to the year 2003 and are
allowable for extraction systems installed prior to the end of the year 1992. Robert F. Hatch of
Cambrian Energy Systems — whose company has been involved in arranging financing for many
U.S. LFG to energy projects - thinks that many of the projects would not be in existence if the tax
credits were not available. Since energy prices are relatively low, some projects today can be
financed only because of the tax credits. The tax credits are intended to help promote the
development of a domestic resource as opposed to using foreign oil (Hatch, 1991). These credits
4-53
-------
have helped to encourage LFG-to-energy projects and also have helped municipalities defray the
cost of environmental regulations.
Table 2. Estimate of Capital Costs for a 1 MWe Landfill Gas-to-Energy Utilization Project
Range of
Cost4
Value
Item
(SI 03)
Percent*
(SlQ3/kWh)b
Extraction/Collection System0
200
13
200-1000
Fees-Plannin g/En vironment/Le gald
30
2
30-1000
Interconnect Cost
76
5
20-500
Generating Equipment
970
65
500-2000
Contingency
m
11
—
Total
1500
100
850-4500
•These costs were provided by Laidlaw Technologies, Inc. (Jansen, 1992)
bAugenstein and Pacey, 1992.
cThe range in cost of gas cleanup systems is $10,000-$500,000/kWh)
dLegal fees are approximately 50% of the total (i.e., -$15,000-$500,000/kWh)
Another barrier to LFG (or waste CH4) utilization can be environmental regulations. Typically
the overall environmental benefit as well as the energy and economic benefits are not considered.
Many LFG projects are operated such that New Source Review is avoided. This results in less
CH4 being utilized. For example, a project in Phoenix, Arizona, is producing 50% of the energy
that is possible to avoid triggering New Source Review. The project was not allowed to take credit
for the offset in emissions from a coal-fired power plant. The developer found out that the New
Source Review process would take an extra 2 years to obtain a permit. Rather than delay receiving
payback on the investment, the decision was made to operate the project at half of its potentiaL
Another dis-incentive is that some operators of LFG-to-energy projects are finding that the cost
of condensate disposal is becoming a major expense. The condensate is formed when the gas is
compressed. The LFG condensate — which is being classified as a hazardous waste — requires
disposal at a Subtitle C facility. This cost [i.e., $0.18/L (~$0.70/gal)] can be significant for a site
where lean-bum engines or turbines are used as compared to the use of flares - where minimal
condensate is collected [i.e., 3,800 L/day (1000 gpd) for lean-bum engines or turbines versus 760
L/day (200 gpd) for flares] (Jansen, 1992).
Industry experts are finding that air, water, and solid waste agencies have conflicting goals.
LFG-to-energy projects have been forced to shut down due to concerns for by-product emissions
of nitrogen oxides (NOx) and carbon monoxide (CO). In California last year, 48 items of state
legislation affecting solid waste were enacted (SWANA, 1992). Regulatory priorities often appear
to conflict.
CONCLUSIONS
Landfills and other waste management sources such as wastewater treatment lagoons, septic
sewage systems, and livestock waste are amenable to cost-effective control and are relatively
significant sources of CH4. The EPA's Global Climate Change Research Program is conducting
research to (1) reduce the uncertainty in global emission estimates for those sources amenable to
4-54
-------
control, (2) target control strategics that are cost-effective, and (3) provide data and information
that will help support regulatory activities and IPCC activities.
Currently, U.S. LFG-to-energy projects recover -1.2 million tonnes of CH4 and produce 344
MWe of power. The proposed CAA regulations for MSW landfill air emissions are expected to
result in additional reductions ranging from 5 to 7 million tonnes of CH4. Utilization of LFG for
those sites affected by the proposed CAA regulations has the potential to result in increased
benefits to the national economy and global environment. The utilization of alternative energy
sources such as "waste" CH4 extends our global fossil fuel resources. Not only are emissions
directly reduced when waste CH4 is recovered and utilized, but emissions are also indirectly
reduced when secondary air emission impacts associated with fossil fuel use are considered
REFERENCES
1. Augenstein, D. and J. Pacey, "Landfill Gas Energy Utilization: Technology Options and Case
Studies," EPA-600/R-92-116 (NTIS PB92-203116), June 1992.
2. Bartone, C. R. 1990. Water Quality and Urbanization in Latin America. Water International,
15: 3-14.
3. Campbell, D., D. Epperson, L. Davis, R. Peer, and W. Gray, "Analysis of Factors Affecting
Methane Gas Recovery from Six Landfills," EPA-600/2-91-055 (NTIS PB92-101351),
September 1991.
4. Federal Register. Vol 56. No. 104. May 30,1991, pp. 24468 - 24528.
5. Hatch, R.F. "The Federal Tax Credit for Non-Conventional Fuels: Its Status and Role in the
Landfill Gas Industry." Proceedings from SWANA's 14th Annual International Landfill Gas
Symposium, 1991.
6. Intergovernmental Panel on Climate Change. "Climate Change - The EPCC Scientific
Assessment." World Meteorological Organization/United Nations Environment Programme.
Edited by J.T. Houghton, G.J. Jenkins, and JJ. Ephraums, 1990.
7. Intergovernmental Panel on Climate Change. "Climate Change 1992 - The Supplementary
Report to the IPCC Scientific Assessment" World Meteorological Organization/United
Nations Environment Programme. Edited by J.T. Houghton, B. A. Callander, and S. K.
Varney, 1992.
8. Jansen. G.R. "The Economics of LFG Projects in the United States." Presented at the
Symposium on LFG/Applications and Opportunities in Melbourne, Australia, February 27,
1992.
9. Khalil, M.A.K. and R.A. Rasmussen, "Constraints on the Global Sources of Methane and an
Analysis of Recent Budgets." Tellus, 42B, 229-236, 1990.
10. Lawson, P. S. Landfill Gas Expert Working Group Summary Report, 1989-1991,
International Energy Agency: Biomass Conversion Agreement: MSW Conversion
Activity, Task VII, AEA-EE-0305, April 1992.
4-E5
-------
11. Organization of Economic Cooperation and Development (OECD). Estimation of
Greenhouse Gas Emissions and Sinks. Final Report from the OECD Experts
Meeting February 1991. Prepared for the IPCC. August 1991.
12. Peer, RJL., DJL. Epperson, D. L. Campbell, and P.V. Brook, "Development of an
Empirical Model of Methane Emissions ftx)m Landfills," EPA-600/R-92-037 (NTIS
PB92-152875), March 1992.
13. Peer, R.L., S.A. Thorneloe, and D.L. Epperson, "A Comparison of Methods for
Estimating Global Methane Emissions from Landfills." Chemosphere, 1992 (In Press).
14. Richards, K.M. Landfill Gas: Working with Gaia, C.A.B. International, Vol. 3, No. 4,
December 1989.
15. Safle, L.M., M.E. Casada, J. W. Woodbury, and K. F. Roos, Global Methane Emissions
from Livestock and Poultry Manure, EPA/400/1-91/048, February 1992.
16. Sandelli, GJ. "Demonstration of Fuel Cells to Recover Energy from Landfill Gas
(Phase I Final Report: Conceptual Study)," EPA-600/R-92-007 (NTIS PB92-
137520), January 1992.
17. Scheepers, MJ.J. "Landfill Gas in the Dutch Perspective," published in Proceedings
of the Third International Landfill Symposium, Sardinia, October 1991.
18. SWAN A. List of Solid Waste Legislation Enacted in 1991.1992.
19. Thorneloe, S.A. Field Test Work for Assessing Greenhouse Gas Emissions from
Wastewater Treatment and Septic Sewage Systems. Memo to Jon Kessler, Air and
Energy Policy Division, Office of Planning and Evaluation, EPA, February 24,1992.
20. Thorneloe, S. A. "Landfill Gas Recovery/Utilization - Options and Economics,"
presented at the Sixteenth Annual Conference by the Institute of Gas Technology
on Energy from Biomass and Wastes, Orlando, FL, March 2-6,1992.
21. Thorneloe, S.A. "Landfill Gas Utilization - Options, Benefits, and Barriers,"
presented at the Second United States Conference on Municipal Solid Waste
Management," Arlington, VA, June 3-5,1992.
22. Thorneloe, S.A. "U.S. EPA's Global Climate Change Program - Landfill Emissions
and Mitigation Research," presented at the Third Internationa] Landfill Symposium,
Cagliari, Italy, October 14,1991.
23. United States Environmental Protection Agency. "Air Emissions from Municipal
Solid Waste Landfills - Background Information for Proposed Standards and
Guidelines." EPA-450/3-90-01 la (NTISPB91-197061), March 1991.
24. United States Environmental Protection Agency, Office of Policy, Planning, and
Evaluation. International Methane Emissions. Draft Report to Congress, July 1992.
4-56
-------
25. United States Environmental Protection Agency, Office of Policy, Planning and
Evaluation. Policy Options for Stabilizing Global Climate. Draft Report to
Congress. February 1989.
26. Valenti, M. "Tapping Landfills for Energy." Mechanical Engineering, Vol. 114,
No. 1, January 1992.
27. Wong, F.P. "Alternative Energy & Regulatory Policy: Till Death Do We Part,"
presented at AWMA Conference on "Cooperative Clean Air Technology -
Advances through Government and Industrial Partnership" in Santa Barbara, CA,
March 21 - April 1,1992.
4-57
-------
Paper 4-E
FUEL CELL POWER PLANT FUELED BY LANDFILL GAS
by: R. J. Spiegel
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
G. J. Sandelli
International Fuel Cells Corporation
South Windsor, CT 06074
ABSTRACT
International Fuel Cells Corporation (IFC), a subsidiary of United Technologies
Corporation, is conducting a U.S. Environmental Protection Agency (U.S. EPA)
sponsored program to demonstrate methane control from landfill gas using a
commercial phosphoric acid fuel cell power plant. This is the world's first commercial-
scale demonstration to control methane emissions from landfills using a fuel cell
energy recovery system. The U.S. EPA is interested in fuel cells for this application
because it is potentially the cleanest energy conversion technology available. This
paper discusses the project in general and describes some results to date, with
emphasis on the landfill gas pretreatment system.
This papa has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication.
4-58
-------
INTRODUCTION
The U.S. EPA has proposed standards and guidelines [1] for the control of air
emissions from municipal solid waste (MSW) landfills. Although not directly controlled
under the proposal, the collection and disposal of waste methane, a significant
contributor to the greenhouse effect, would result from the emission regulations. This
U.S. EPA action will provide an opportunity for methane control as well as for energy
recovery from the waste methane that could further benefit the environment. Energy
produced from landfill gas could offset the use of foreign oil, and air emissions
affecting global warming, acid rain, and other health and environmental issues.
To demonstrate that methane control and subsequent energy recovery via fuel
cells are technically, economically, and environmentally feasible, two key issues must be
addressed: to define a gas pretreatment system to render the landfill gas suitable
for fuel cell uses and to design the modifications necessary to ensure that rated
power is achieved from the dilute methane fuel. Only relatively simple engineering
modifications, albeit initially costly to implement, are required to ensure that rated
power is achieved from the dilute landfill gas. The toughest and most critical problem
is the gas cleanup system. Therefore, the R&D focus of this project is on the landfill
gas contamination problem.
International Fuel Cells Corporation (IFC), a subsidiary of United Technologies
Corporation, was awarded a contract by the U.S. EPA to demonstrate methane
destruction and energy recovery from landfill gas using a commercial phosphoric acid
fuel cell. IFC is conducting a three-phase program to show that fuel cell technology is
economically and environmentally feasible in commercial operation. Work was initiated
in January 1991. A U.S. EPA report [2] describes the results of Phase I, a conceptual
design, cost, and evaluation study, which addresses the problems associated with
landfill gas as the feedstock for fuel cell operation.
Phase II of the program includes the design, construction, and testing of the
landfill gas pretreatment module to be used in the demonstration. Its objective will be
to determine the effectiveness of the pretreatment system design to remove critical
fuel cell catalyst poisons such as sulfur and haiides. A challenge test is planned to
show the feasibility of using the pretreatment process at any landfill in conjunction
with the fuel cell energy recovery concept. The gas pretreater is described here.
Phase ill of this program will be a demonstration of the fuel cell concept. The
demonstrator will operate at Penrose Station, an existing landfill-gas-to-energy facility
owned by Pacific Energy in Sun Valley, CA. Penrose Station is an 8.9 MW internal
combustion engine facility supplied with landfill gas from four landfills. The electricity
produced by the demonstration will be sold to the electric utility grid.
4-59
-------
Phase II activities began in September 1991, and Phase III activities are
scheduled to begin in June 1993.
FUEL CELL CONCEPT
During Phase I, a commercial fuel cell methane destruction and energy recovery
system concept was designed. The system, shown in Figure 1, is based on
commercially available equipment adapted for operation on landfill gas. The system
was sized to be broadly applicable to a large number of landfills.
PENROSE
STATION
GAS WELLS
AND
COLLECTION
SYSTEM
(PACIFIC ENERGY)
UTILITY
POWER
UNE8
AC POWER r
TO GRID
PCS!
FUEL CELL
POWER
PUNT
(ONSI CORP.)
OAS-GUARD •
OAS PRCTREAT
KENT SYSTEM
(BIOGAS
DEVELOPMENT
INC.)
COGENERATION
HEAT
LANDFILL
NATURAL GAS
SOUTHERN CALIFORNIA GAS COMPANY
Figure 1. Project Conceptual Design
Landfill gas is collected by a series of wells in the MSW landfill and piped to a
gas pretreatment module. The pretreatment module removes contaminants such as
sulfur and halides which affect the operation of the fuel cells. The contaminants are
concentrated on an absorption bed to a predetermined level. Then during a
regeneration cycle they are stripped from the absorption media and destroyed by
incineration. Hydrocarbon condensates which form in the pretreater are also
incinerated. The resulting output is a medium heat value methane fuel suitable for use
in the fuel cell.
4-60
-------
The concept utilizes four modular 200-kW phosphoric acid fuel cells generating
electricity to be sold tc the electric utility grid. The fuel cell power plants are
adaptations from the natural-gas-fueled PC25 fuel cell sold by ONSI Corporation, an
IFC subsidiary. Only simple modifications are required to ensure that rated power is
achieved from the dilute landfill gas.
The MSW landfills in the United States were evaluated to determine the potential
commercial market which could be derived using a 200-kW fuel cell. Each fuel cell
would consume 2832 scmd (100,000 scfd) of landfill gas to generate 200 kW,
assuming a heating value of 19 MJ/dscm (500 Btu/dscf).
The potential commercial market available'for fuel, cell operation was evaluated
using a U.S. EPA estimate of methane emissions in the year 1997 [3] and an estimate
of landfill gas production rate of 0.0028 scm/yr per tonne (0.1 scf/yr per ton) of
refuse in place [4]. An estimated 4370 MW of power could be generated from the
7480 existing and closed sites that were identified [2]. The largest number of
potential sites greater than 200 kW occurs in the 400 to 1000 kW range. This
segment represents a market of 1700 sites or 1010 MW. The assessment concluded
that these sites are ideally suited to the fuel cell concept. The concept can provide a
generating capacity tailored to the site because of the modular nature of the
commercial fuel cell. Sites in this range are also less well served by competing
options, especially Rankine and Brayton Cycles, which exhibit poorer emission
characteristics at these power ratings.
ENVIRONMENTAL ASSESSMENT OF THE FUEL CELL SYSTEM
The environmental impact from commercial application of the fuel cell concept
to the market described previously can be assessed. For the purpose of the
evaluation, a site capable of supporting four fuel cell power modules (800 kW total
capacity) was selected. The site would produce approximately 11,328 scmd
(400,000 scfd) of landfill gas per day. The gas contains approximately 50% methane
with a heating value of 19 MJ/dscm (500 Btu/dscf).
The analysis of the environmental impact shows that the fuel cell can be
designed to eliminate the methane and non-methane organic compounds (NMOCs)
from landfill gas streams. With the fuel cell system, significant amounts of CO2 and
S02 will also be reduced due to the fuel cell energy generation. Using an 80% capacity
factor for the fuel cell and offsetting emissions from electric utility power generation
using a coal-fired plant meeting New Source Performance Standards, it can be shown
that for the example site the fuel cell energy conversion system provides 5.6 million
kWh of electricity per year and a reduction of emissions for CH4, NMOCs, C02, SO2,
and CO of 1200, 35, 4200, 36, and 0.6 Mg/yr, respectively. These reductions can be
4-61
-------
used as environmental offsets, particularly in critical areas such as California or other
locations with severe environmental restrictions.
The environmental impact of the fuel cell concept to the potential U.S. market
can also be estimated. If the viable market is assumed to range in sites that have
energy capacities from 200 kW up to 1 MW, then the fuel cell system can provide an
approximate net reduction in emissions for CH4, NMOCs, CO2.SO2, and CO of 2x10s,
6x104, 7x106, 6x104, and 1x103 Mg/yr, respectively.
FUEL CELL POWER PUNT
A design of a fuel cell power plant was established to identify the design
requirements which allow optimum operation on landfill gas. Three issues specific to
landfill gas operation were identified which reflect a departure from a design
optimized for operation on natural gas. A primary issue is to protect the fuel cell
from sulfur and halide compounds not scrubbed from the gas in the fuel pretreatment
system. An absorbent bed was incorporated into the fuel cell fuel preprocessor
design which contains both sulfur and halide absorbent catalysts. A second issue is
to provide mechanical components in the reactant gas supply systems to
accommodate the larger flow rates that result from use of dilute methane fuel. The
third issue is an increase in the heat rate of the power plant by approximately 10%
above that anticipated from operation on natural gas. This is a result of the
inefficiency of using the dilute methane fuel. The inefficiency results in an increase in
heat recoverable from the power plant. Because the effective fuel cost is relatively
low, this decrease in power plant efficiency will not have a significant impact on the
overall power plant economics.
The landfill gas power plant design provides a packaged, truck transportable,
self-contained fuel cell power plant with a continuous electrical rating of 200 kW. It is
designed for automatic, unattended operation, and can be remotely monitored. It
can power electrical loads either in parallel with or isolated from the utility grid.
In summary, a landfill-gas-fueled power plant can be designed to provide 200
kW of electric output without need for technology developments. The design would
require selected components to increase reactant flow rates with a minimum pressure
drop. To implement the design would require non-recurring expenses for system and
component design, verification testing of the new components, and system testing to
verify the power plant performance and overall system integration.
4-62
-------
LANDFILL GAS PRETREATMENT SYSTEM
The available information on landfill gas compositions was evaluated to
determine the range of gas characteristics which a fuel cell landfill-gas-to-energy
power plant will encounter. This information was used to set the requirements for the
gas pretreatment and fuel cell power plant designs.
The major non-methane constituent of landfill gas is CO2. The CO2 ranges from
40 to 55% by volume of the gas composition with a typical value of 50%. Other
diluent gases include nitrogen and oxygen, which are indicative of air incursion into the
well (most frequently in perimeter wells). Nitrogen concentrations can range as high
as 15%, but typical values are 5% or less. Oxygen concentrations are monitored
closely and held low for safety reasons.
Landfill gas constituent compounds reported by U.S. EPA [3] indicate a typical
value for the total NMOCs of 2700 ppmv, ranging from 240 to 14,000 ppmv
(expressed as hexane). The NMOC concentration in the landfill gas is an important
measure of the total capacity required in the gas pretreatment system, while the
specific individual analyses provide a basis for gas pretreatment subcomponent sizing.
The specific contaminants in the landfill gas, of interest to the fuel cell, are sulfur and
halides (chiefly chlorides and fluorides). The sulfur level ranges from 1 to 700 ppmv,
with a typical value on the order of 21 ppmv. Sufficient data were not available to
assess the range of the halides, but a typical value of 132 ppmv was calculated for
this contaminant [3]. The range of contaminant values varies not only from site to
site, but also at any given site with time due to seasonal weather or moisture
content. These characteristics require the pretreatment system design to be capable
of handling these gas quality variations to avoid expensive site specific engineering of
the pretreatment design which would affect the marketability and economics of the
concept.
Figure 2 illustrates the overall design strategy for the gas cleanup system. As
shown, the raw landfill gas pretreatment system is designed to reduce the primary
fuel cell contaminants (sulfur and halides) to levels between 1 and 10 ppm. A nominal
value of around 3 ppm is the design goal. An additional spool piece is added to
protect the fuel cell from sulfur and halide compounds not removed from the gas in
the pretreatment system. This is an absorbent bed incorporated into the fuel cell
preprocessor design which contains both sulfur and halide absorbent catalysts. Note
that the capacity of this bed for contaminant removal is dictated by the number of
hours the fuel cell can operate before these poisons contaminate the fuel cell, requiring
refurbishing.
4-63
-------
Raw w
LFG
Pretreatment
system
Spool
piece
PC25
Pretreat capability PC2S requires
1-10 ppm 0.15 ppm for 8000 hours
0.01 ppm for 40,000 hours
LFG ¦ Landfill Gas
Figure 2. Block Diagram of Overall Landfill Gas Contaminant Removal Process
A block diagram of the landfill gas pretreatment system is shown in Figure 3.
The landfill gas pretreatment system, designed by Bio-Gas Development, Inc., is
optimized to handle a wide range of landfill gas contaminant levels and compositions.
This was achieved by utilizing a staged contaminant removal approach which
enhances the operation of each successive process step. This is accomplished by
removal of contaminants which adversely affect downstream processes and
controlling the temperature of each step to optimize its efficiency. Figure 3
summarizes the staged contaminant removal processes which produce clean landfill
gas to meet the fuel cell specification.
fttaENCMLHONOU
rtlWT ITASI
MFNtOCIUtSa
CONSUMM
¦ranwuaia
MIX. IIIYI
DtMTMATIQM
1
T
>
HOCK IRABU
icmtTU
CAM SON
VOCREM0VM.
RiaMMM*
mlmvi
MWWML
TO
PUCl
X
I
r«
i
i
x
MCUMMIOR
MAT
MCOWmT
rat
COWOtMMTt
MPOMZATION
MaiMCMIISM
MATIN*
Figure 3. Block Diagram of the Landfill Gas Staged Contaminant Removal System
4-64
-------
The landfill gas is first cooled to approximately 1°C to remove water and heavy
hydrocarbon contaminants. Following further dehydration in absorption media, the
landfill gas temperature is lowered to optimize the operating temperature of the
downstream activated carbon/NMOC removal and molecular sieve/H2S removal
steps. We anticipate that low molecular weight NMOCs will be removed in the second
cooler by non-steady state adsorption in the liquid film on the heat exchanger tubes, a
concept proprietary to Bio-Gas Development, Inc. The final stage of NMOC removal is
by absorption in activated carbon bed media. Hydrogen sulfide is selectively
absorbed on a molecular sieve media placed downstream of the activated carbon in
the same vessel.
The presence of CO2 in the gas produces competition for active adsorption
sites in the H2S mole sieve and can potentially produce carbonyl sulfide (COS) by
reaction with H2S. This competition is minimized by operating at low pressure which
favors H2S adsorption over C02 adsorption. The production of COS is minimized by
maintaining low bed temperatures to slow the kinetics for this reaction. The specific
adsorption media was selected to maximize H2S removal.
Lastly a particulate filter will remove fines which may be produced from
successive thermal regeneration cycling of the adsorption beds. The cleaned landfill
gas is delivered to the fuel cell.
Approximately 15% of the cleaned gas is used to regenerate the absorption
beds. This gas, containing a high concentration of desorbed contaminants, is flared
to achieve 98% destruction of NMOCs.
CONCLUSIONS
A demonstration project design was established which addresses the key
technical issues facing commercial application of the fuel cell methane control and
energy recovery concept to the market. A site was selected (Penrose Power Station)
which represents the landfill gas market. A gas pretreatment system has been
designed, and construction of the system is underway. No technical "show stoppers"
are apparent, but the success of the project clearly will be determined by the
effectiveness of the landfill gas pretreatment system to remove critical fuel cell
catalyst poisons. These critical tests, which will commence soon, will ultimately decide
the fate of fuel cell application on landfill gas.
REFERENCES
1. U.S. Federal Register, May 30, 1991, Part III Environmental Protection Agency,
40 CFR Parts 51, 52 and 60; Standards of Performance for New Stationary
4-65
-------
Sources and Guidelines for Control of Existing Sources; Municipal Solid Waste
Landfills; Proposed Rule, Guideline and Notice of Public Hearing.
2. Sandelli, G.J., Demonstration of Fuel Cells to Recover Energy From Landfill Gas
(Phase I Final Report: Conceptual Study), EPA-600/R-92-007 (NTIS PB92-
137520), January 1992.
3. Air Emissions from Municipal Solid Waste Landfills - Background Information for
Proposed Standards and Guidelines, EPA-450/3-90-0l1a (NTIS P091 -197061).
March 1991, p. 3-30.
4. Maxwell, Greg, "Will Gas-To-Energy Work at Your Landfill?," Solid Waste &
Power, June 1990, p.44.
4-66
-------
METHANE EMISSIONS FROM RICE AGRICULTURE
M.A.K. Khalil
M.J. Shearer
R.A. Rasmussen
Global Change Research Center
Oregon Graduate Institute
Beaverton, Oregon 97006 U.S.A.
ABSTRACT
Rice agriculture has long been recognized as a major source of methane (CH«) .
Global budgets of methane have generally included emissions of about 100
Tg/yr (Tg - 10u grams) from rice agriculture (range of 50-300 Tg/yr) and
constituting about 20% of emissions from all sources (range 14%-60%).
During the last decade a number of systematic experiments have been reported
on methane emissions from rice fields. Seasonal averages range from 0 to 40
mg/m2/hr. Factors affecting the flux of methane include irrigation regime,
fertilizer, soil temperature, and soil type.
The most recent global estimates put the emissions from rice paddies at 50 to
100 Tg/yr. The major cause of increasing methane emissions from rice paddies
over the past 50 years appears to be the tremendous increase in area planted
to rice. Emissions appear to have stabilized over the past decade. Future
increases in methane emissions from rice will probably depend on access to
irrigation and the use of organic fertilizer.
The work described in this paper was not funded by the U.S. Environmental Protection Agency. Hie
contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred.
4-67
-------
1. INTRODUCTION
Rice agriculture has long been recognized as a major source of methane (CH4).
Global budgets of methane have generally included emissions of about 100 Tg/yr from
rice agriculture (range of 50-300 Tg/yr) and constituting about 20% of emissions
from all sources (range 14ft-40%) (Ehhalt and Schmidt 1978; Donahue, 1979; Khalil &
Rasmussen 1983; Blake 1984; Bolle et al. 1986; Bingemer and Crutzen 1987; Cicerone
and Oremland 1988; Warneck 1988). The increase in rice agriculture was probably one
of the main contributors to the increase of methane during the last century. It is
estimated that rice agriculture contributes some 300 ppbv of methane to the present
atmosphere and may be responsible for some 20% of the increase of methane during the
last century (Khalil and Rasmussen, 1991).
Estimating the flux'of methane from rice fields in various parts of the world
requires knowledge of two factors: the emission rates and the regional or global
extrapolant. The emission rate or flux depends on different internal and external
variables. Internal variables include: soil characteristics; rice cultivar; and
soil microbiology. External factors include: soil temperature (driven by solar
radiation); meteorological conditions; water level, which is affected by rainfall
and availability of irrigation; and treatments such as the type and amount of
fertilizers. The extrapolant is the area with similar values of the different
variables. The flux of CH, from rice fields from any region is:
where $ is the flux and Aj are the areas within the region of interest that have
similar values of the variable x,.
At present most integrated regional and global emissions are calculated using a
measured flux from a region and multiplying it by the area of that region in the
hope that the local measurement represents an average emission rate for the whole
area. In recent years some progress has been made in identifying the factors (x,)
that can be used to obtain better regional flux estimates.
2.1 FLUX MEASUREMENTS
During the last decade a number of systematic experiments have been reported on
methane emissions from rice fields. All are based on static chamber methods. While
there are many variants, the method consists of enclosing a part of the rice fields
within a chamber and taking periodic samples. The samples are analyzed for methane
content usually by gas chromatography using flame ionization detectors (GC/FID).
Methane emitted from the rice builds up in the chamber. The rate of accumulation
is directly proportional to the flux or the rate of emission from the area covered
by the chamber. The relationship is;
N
(1)
2. METHANE EMISSION RATES FROM RICE FIELDS
4-68
-------
Flux = PJU x 10"* *£ (2)
No A dt
where p is the density of air (molecules/m3) , C is the concentration of methane
(ppbv) and dC/dt is in (ppbv/hr), A is the area covered
-------
Table 2.1: Melhane Flux Measurements
Study
Location
Treatment
Rice Cultivar
Soil Type
Flux mg/mVhr
Emission
Season days
Cicerone et al., 1983
California, USA
Nitrogen fertilizer
M101
Vcrtisol, Capay clay
10.4
100
Seiler el al., 1984
Andalusia, Spain
Nitrogen fertilizer
Bahfa
Not given
4.0
125
SchUtz et al., 1989
Vercel|i, Italy
Unfertilized
Rice straw
Nitrogen
Straw + N
Roma
Sandy Ijoam
11.7
17.6
10.5
18.1
105 to 120
Yagi and Minami, 1990
Ibaraki Prefecture,
Japan
Mineral(avg)
Organic(avg)
Rice straw
Mineral(avg)
Organic(avg)
Mineral
Rice straw
Koshihikari
Gley soil
Peat soil
Andosol: humic
n
Andosol: volcanic
M
2.9
6.7
16.3
1.3
3.1
0.2
0.4
119
115
115
124
129
125
115
Sass et al., 1990, 19918, and
1991b
Texas, USA
Nitrogen
Rice straw
Nitrogen
Rice Straw
Nitrogen
Rice Straw
Jasmine 85
Typic Pelludert
n
linlic Pelludert
it
Vertic Ochraqualf
ft
8.7
15.2
2.5
5.6
14.0
18.2
85
86
85
86
76 to 85
Dai, 1988; and SchUtz et al.,
1990
Hangzhdu, China
Late rice, average
Early rice, average
Not given
Not given
26.4*
66*
62
70
Khalil et al., 1991
Sichuan, China
Organic
local and hybrid
"purple soil"
36.6
120
Chen et al., 1992
Beijing, China
Nanjing, China
N + Organic F*
N + Organic: II
N + Organic: D
Average, all
Huang jinguang
Shanyou 63
Sandy Loam
"yellow-brown
earth"
34.1
14.6
-00
. 88
* Values for SchUtz et al. (1990) were digitized from the figures.
* F: flood irrigation; 11: intermittant irrigation; D: dry culture.
-------
populations and ultimately the methane flux from the paddy fields.
Properties which affect methane flux can be roughly divided into factors internal
to the plant and planting site, and external factors such as the weather or man's
actions. These can be re-organized for the purpose of modelling into categories of
factors which change with time or space.
Table 2.2 Factors affecting methane fluxes from rice paddies.
Internal
External
Temporal
Spatial
Factors
Factors
Factors
Factors
Soil:
Weather:
Season
Type of
Texture
Temperature
Fertilizer
Mineralogy
Rainfall
Planting date
Eh/pH buffer system
Irrigation
Agricultural
projects
Microbiology
Practices:
Irrigation
Soil Type
Rice Cultivar
Fertilization
The internal factors of soil microbiology, soil properties, and different rice
cultivars are at present the most difficult to incorporate into regional estimate
of methane flux from rice paddies. The variation in the factors is very high and
their effects can be difficult to measure in the field.
External factors are sometimes easier to measure, but the relationships vary from
year to year and place to place. For example, Khalil et al. (1991) found a good
correlation between soil temperature and methane flux, with a Qw of -3 (see Figure
1). However, they found a slightly different relationship between 1988 and 1989;
and Chen et al. (1992) found higher average fluxes in Beijing vs. Nanjing, China,
with lower average temperatures.
Rice agriculture may be roughly divided into wetland and dryland (or upland) culture
(Grist, 1986; Neue et al., 1990). In wetland culture, the soil is prepared
(puddling) to reduce water loss, and dikes or levees are built to contain the water.
Wetland culture may be separated into irrigated and rainfed riceT While rainfed
rice is planted during the wet season, it is susceptible to drought, and if the soil
dries, methane emissions stop. Dryland culture is a low-yield, subsistence
agriculture, highly susceptible to drought. Dryland rice may not be a source of
methane at all, as the soil is not saturated long enough for a methanogen population
to build up. Most of the studies in Table 2.1 were carried out where flood
irrigation was practiced. The exception, Chen et al. (1992), tested several
irrigation regimes and found lower fluxes whenever the top of the soil was allowed
to dry. Field measurement on wetlands and on tropical soils (Harris, 1982; Keller,
1986) found that methane flux stops and methane oxidation (conversion to C02) begins
once the soil dries out.
4-71
-------
a
•s.
m
E
M
E
H
P
100
80
60
40
20
16 20 24 28
Temperature (C>
32
¦ 1988
• 1989
Figure 1. Relationship between flux and soil temperature. (From Khalil and
Rasmussen, ES&T, 1991)
3. REGIONAL AND GLOBAL EXTRAPOLATIONS
Every methane budget includes an estimate of emissions from rice paddies; however,
only two have been published that include details of regional emissions (Aselmann
and Crutzen, 1989, hereafter Aselmann and Crutzen; Matthews et al., 1991, hereafter
Matthews et al.). While both studies were done to provide data for global
transport-chemistry models for methane, they concentrated on different variables,
— and'th* authors made different types of- information available. An -estimate of
regional emissions by the Global Change Research Center (GCRC) at the Oregon
Graduate Institute is compared to these two studies.
3.1 RECENT STUDIES AND DATA BASES
Aselmann and Crutzen provided detailed tables of the percent of the area in 2.5*
latitude by 5* longitude boxes. The percentages reflect the area in rainfed and
irrigated rice only for Asia (no upland or dryland rice), but show total area for
Africa, Central and South America. Fluxes were adjusted from the work of Holzapfel-
Pschorn and Seiler (1986), taking a base rate of 300 mg/m2-day (12.5 mg/m2-hr) for
average soil temperatures of 20°C and below, and assuming a linear relationship
between flux and temperature up to 1000 mg/mJ-day (42 mg/m2-hr) for soil temperatures
of 30*C. The reader must construct the exact fluxes used, as well as the monthly
.4-72
-------
allocations of area under cultivation.
Matthews et al. published detailed rice crop calendars, indicating the months of
possible cultivation of rice by country, each Indian state and each province of
China. Area was allocated by 1" latitude by 1* longitude cells. Matthews ec al.
used a flux rate of 0.5 g/m2-day (21 mg/m2-hr) for all areas.
For our (GCRC) estimates we developed an inventory of direct flux measurements and
modified the information from Matthews et al. regarding areas and growing seasons
to estimate global annual emission rates. The flux from rice fields is calculated
using Equation (1), where the Xj are the length of the season of methane emission,
the area in rice (irrigated and rainfed), and the seasonal (or monthly) average flux
of methane (mg/mJ-hr) . Each of these main variables is discussed separately.
3.2 SEASON LENGTH
The estimate of the season length from the literature may be confused by whether
season refers to the season of methane emission; the cultivar growing season (or
time from planting or transplanting to harvest); the total growing season (or frost
free season); the total growing duration, which in the case of transplanted rice,
includes the time in the seedling beds. For example, in the study of Sass et al.
(1991a) the total growing season is about 245 days, the rice cultivar needs 140 days
from planting to harvest, while methane was emitted 85 days (flood irrigated
period).
Growing season was estimated from Matthews et al., Tables la-c. The growing season
was defined to be the length of time from first seeding or transplanting until
harvest. Table 3.1 compares the growing season and season of methane emission
estimated from the literature (see Table 2.1) with the growing season estimated from
Matthews et al. With the exception of Italy, the difference between the estimated
growing season and season of methane emission is about 40 to 50 days.
Table 3.1: Seasonal Factors for Methane Emissions From Rice Fields
References
Planting
Season
days
CH« Season
days
Growing
Season
days*
Cicerone et al. (1983): USA
145
100
152
Holzapfel-Pschorn and Seiler
(1986): Italy
147
126
122
Yagi et al. (1990): Japan
140
115
152
Sass et al. (1991a): USA
140
85
152
Khalil et al. (1991): Sichuan
120
120
168
Chen et al. (1992): Beijing
100
87
137
+ Estimated from Matthews et al. (1991). See text.
4-73
-------
We assumed chat using che growing season calculated from Che crop calendar and
average flux rates reported in the literature will probably lead to an overestimate
of total emissions. Therefore, Matthews et al.'s growing seasons were modified as
follows: growing seasons of 140 to 170 days were reduced by 30 days; growing seasons
of 110 to 140 days were reduced by 20 days; and growing seasons of fewer than 110
days were reduced by 10 days. This adjustment is arbitrary, but better reflects
results of che field measurements.
3.3 METHANE FLUX FACTORS
Fluxes from the studies listed in Table 2.1 were used for che major seasonal
divisions of Chinese and Indian rice; the countries where samples were taken, and
estimated for other rice growing countries.
3.4 AREA PLANTED TO RICE
Time series of harvested rice area were constructed for all rice growing countries.
Historical statistics compiled by Mitchell (1980, 1982 and 1983), and more recent
data available from the Food and Agriculture Organization of the United Nations
(Yearbook, various years between 1956 and 1988) were the primary sources of
information. For China, the areas used were reported by either the USDA (1984) or
the 1985 China Agriculture Yearbook. From 1980 to 1990 the total areas are from the
FAO Production Yearbooks or the China Agriculture Yearbooks, which are in good
agreement for that period.
Total area was broken down further by province for China and by state for India.
The fraction of the total areas in each Chinese province was calculated by taking
the average of the areas reported in the China Agriculture Yearbooks (all years) and
the values in Matthews et al., Table lc. The fraction of the total area in each
Indian state was calculated using the values given in Table lb of Matthews et al.
and Appendix 9 in Bansil (1984). These percentages represent only the 1960s;
however, the areas which are the largest producers are probably the same.
The total areas were reduced by the percentage in dryland (upland) rice culture.
Estimates of the percentages were taken from tables in Huke (1982), Grist (1986),
Morris et al. (1984), and De Datta (1975). Where there was more than one estimate
per country, an average was taken. If there was no estimate for a particular
country, but there was reason to suppose a significant percentage of the rice area
was in upland rice, e.g. countries in Africa or Central America, an average was
taken of the published estimates for neighboring countries, or the average for the
continent was used, usually for African nations.
4. REGIONAL AND GLOBAL EMISSIONS
Different estimates of global and regional emissions are compared in this section.
Recent estimates of the global source range from 50 to 100 Tg (1012 grams) per year
4-74
-------
(see Table 4.1). Source estimates by country vary greatly with the assumptions made
on the importance of different factors affecting the methane flux, and the
information on the factors presently available.
4.1 GLOBAL EMISSIONS
Three recent estimates of the global rice source are shown in Table 4.1. (Matthews
et al. (1991) assumed a global source of 100 Tg.)
GCRC
Aselmann and
IPCC
(1992)
Crutzen (1989)
(1992)
Tg/year
80
92 (53*)
60
Year of Estimate
1990
1985
Not given
T Number in parenthesis
assumes a
constant flux of 13 mg/mJ-hr
^0 degree latitude band
Matthews et a I .^BGCRC ^Ba&(
Figure 2. Methane from rice paddies apportioned to 10 degree latitude bands: a
comparison of three studies.
4-75
a.
-------
The major difference between these estimates is probably in the way a weighted
average flux rate is calculated. Bachelet and Neue (in press) calculated a range
of 40 to 80 Tg/year in the rice source for Asia', making different assumptions about
the effects of soil type, rice yield, temperature, and organic matter incorporated
in the soil. The Global Change Research Center (GCRC) estimate takes a seasonally
averaged flux, based on field studies, and applies it to the seasonal area in
wetland rice culture.
4.2 REGIONAL EMISSIONS
Regional emissions are compared by latitudinal bands, and by countries. The
methane source, apportioned to 10 degree latitude bands, is shown in figure 2. The
overall similarity between the estimates is the result of the allocation of rice
area; all three studies use the FAO Agricultural Yearbook data. The GCRC estimate
and Aselmann and Crutzen (1989) differ from Matthews et al. (1991) in the 20 to 30"
N latitude band because the former studies reduce the total area by the estimated
area in dryland cultivation. The GCRC estimate also reduces rice area in Africa and
entral and South America by the percent in dryland cultivation.
v.
(D
-
2?
n
a>
2
cj>
90
ED
60
50
AO
h
f\.r
- if i
ly— 1
1
2.5
1.5
0.5
-0.5
CM
<
(D
0)
>-
a>
c
10
r
G>
1900 • 1910 1920 1930 19«0 1950 1960 1970 19B0 1990
Year -
_»_Rate of Change of Rice Paddy Source
Global Methane Source from Rice
Figure 3. The GCRC time series of global CH« from rice and rate of change of
global rice emissions
4-76
-------
5. THE ROLE OF RICE AGRICULTURE IN THE BUDGET OF ATMOSPHERIC METHANE
The increase in rice agriculture was likely one of the main contributors to the
increase of methane during the last century. It is estimated that rice agriculture
contributes some 300 ppbv of methane to the present atmosphere and may be
responsible for some 20% of the increase of methane during the last century (Khalil
and Rasmussen, 1991). However, factors which were important causes of the increase
in the past are changing, and probably will not be as important in the future (see
figure 3). The GCRC time series of global methane emissions from rice is largely
influenced by tremendous increase in area planted to rice in the last four decades.
By the early 1980s this growth had slowed significantly. A linear projection of the
increase of methane from rice made from the trend in the last decade would indicate
an increase of only 0.2 Tg/year.
The work described in this paper was not funded by the U.S. Environmental Protection
Agency and therefore the contents do not necessarily reflect the views of the Agency
and no official endorsement should be inferred.
6. ACKNOWLEDGEMENTS
Support for this work was supplied by grants from the U.S. Department of Energy
(#DE-FG06-85ER6031) and the National Science Foundation (ATM-8414020). Additional
support was provided by Andarz Co. (#EPA 75-110-01).
7. REFERENCES
1. Aselmann, I., and P.J. Crutzen. Global Distribution of Natural Freshwater
Wetlands and Rice Paddies, their Net Primary Productivity, Seasonality and
Possible Wetland Emissions. J. of Atmos. Chen. 8, 307-358, 1989.
2. Bachelet, D., and H.U. Neue. Methane Emissions from Wetland Rice Areas of
•Asia. Ghefflosphere. *Special-Issue-on .Atmospheric. Methane. La press. 1992. • •
3. Bansil, P.C. Agricultural Statistics in India. A Guide. Third Rev. Edition.
Oxford & IBH Publishing Co., New Delhi, India, 1984.
4. Bingemer, H.G. and P.J.Crutzen. The production of methane from solid wastes
J. Geoohvs. Res. 92, 2181-2187, 1987.
5. Blake, D.R. Increasing concentrations of atmospheric methane (Ph.D
dissertation, Univ. of California at Irvine), 1984. 213 pp.
6. Bolle, H.-J., W. Seiler and B.Bolin (1986), Other greenhouse gases and
aerosols. In: The Greenhouse Effect, Climate Change and Ecosystems (SCOPE 29)
J. Wiley and Sons, N.Y., 1986. 157-198.
5. Boone, D.R. Biological Formation of Methane. In: Global Atmospheric Methane.
-4-77
-------
Ed.: K.A.K. Khalil and H.J. Shearer. NATO ARW Series, in press.
6. Chen, Z. , L. Debo, K. Shao, and B. Wang. Features of CH4 Emission from Rice
Paddy Fields in Beijing and Nanjing, China. Chemoschere. Special Issue on
Atmospheric Methane, in press, 1992.
7. China Agriculture Yearbook 1985, 1986, 1987, 1989. Agribookstore, Hampton,
VA.
8. Cicerone, R.J., J.D. Shetter, and C.C. Delviche. Seasonal Variation of
Methane Flux from a California Rice Faddy. J. of Geophvs. Res.. 88(C15),
11,022-11,024, 1983.
9. Cicerone, R.J. and R.S.Oremland. Biogeochemical aspects of atmospheric
methane. Global Biogeocheniical Cycles 2. 299-327, 1988.
10. Dai, Aiguo. CH« Emission Rates from Rice Paddies in HangZhou China During
Fall of 1987. M.S. Thesis, Institute of Atmospheric Physics, Academia
Sinica, Beijing, P.R.C., 1988.
11. De Dacta, Surajit K. Upland Rice Around the World. In: Major Research in
Upland Rice. International Rice Research Institute, Los Banos, Phillipines,
1975.
12. Donahue, T.M. The atmospheric methane budget. In: Proceedings of the NATO
Advanced Study Institute on Atmospheric Ozone: Its Variation and Human
Influences, A.C. Aikin, Ed. U.S. Department of Transportation, Washington
D.C, 1979.
13. Ehhalt, D. and U.Schmidt. Sources and sinks of atmospheric methane. PAGEOPH
116, 452-464, 1978.
14. Grist, D.H. Rice. Longman, Inc., New York, U.S.A. 6* Edition, 1986.
15. Harris, R.C. , D.I. Sebacher, and F.P. Day, Jr. Methane Flux in the Great
Dismal Swamp. Nature. 297, 673-674, 1982.
16. Holzapfel-Pschorn, A. and W. Seiler. Methane Emission During a Cultivation
Period From an Italian Rice Paddy. J. of Geophvs. Res.. 91(D11), 11,803-
11,814, 1986.
17. Huke, R.E. Rice Area bv Tvne of Culture: South. Southeast, and East Asia.
International Rice Research Institute (IRRI), Los Banos, Phillipines, 1982.
18. IFCC. Climate Chance 1992. The Supplementary Report to The IPCC Scientific
Assessment. Ed. J.T. Houghton, B.A. Callander, and S.K. Varney. Cambridge
University Press, G.B., 1992.
19. Keller, M., W.A. Kaplan, and S.C. Wofsy. Emissions of NjO, CH, and C02 From
Tropical Forest Soils. J.Geophvs. Res.. 91(D11), 11791-11802, 1986.
20. Khalil, M.A.K. and R.A.Rasmussen. Sources, sinks and seasonal cycles of
- 4-78
-------
atmospheric methane, J. Geophvs. Res. 88. 5131-5144, 1983.
21. Khalll, M.A.K. and R.A. Rasmussen. The Global Methane Cycle. In:
Proceedings of the Indo-US Workshop on Impact of Global Climatic Changes on
Photosythesis and Plant Productivity, Hew Delhi, India, Jan. 8-12, 1991.
22. Khalil, M.A.K., R.A. Rasmussen, M.-X. Wang and L. Ren. Methane Emissions
from Rice Fields in China. Env. Sci. and Tech.. 25, 979-981, 1991.
23. Matthews, Elaine, Inez Fung, and Jean Lerner. Methane Emission from Rice
Cultivation: Geographic and Seasonal Distribution of Cultivated Areas and
Emissions. Global Biofeochemical Cycles. 5(1), 3-24, 1991.
24. Mitchell, B.R. European Historical Statistics. 2nd Rev. Ed. Facts on File,
New York, U.S.A., 1980.
25. Mitchell, B.R. International Historical Statistics. Africa and Asia. New
York University Press, 1982.
26. Mitchell, B.R. International Historical SWiStjCS. The Americas and
Australasia. Gale Research Co., Detroit, Michigan, 1983.
27. Morris, R.A., D.J. Greenland, and R.E. Huke. Remote Sensing and Rice
Research. In: Applications of Remote Sensing for Rice Production. 33-54.
Eds. Ardash Deepak and K.R. Rao. A. Deepak Publishing, Hampton, VA, U.S.A.,
1984.
28. Neue, H.-U., P. Becker-Heidmann, and H.W. Scharpenseel. Organic Matter
Dynamics, Soil Properties, and Cultural Practices in Rice Lands and their
Relationship to Methane Production. Xq: Soils and the Greenhouse Effect.
Ed. A.F. Bouwman. J. Wiley and Sons, N.Y., 1990.
29. Neue, H.-U., and P.A. Roger. Rice Agriculture: Factors Controlling Emissions 1
In: Global Atmospheric Methane. Ed. M.A.K. Khalil and M.J. Shearer. NATO ARtf
Series, in press.
30. S«ss, -R-.L: ,-F.M. Wisher; PrAt-Harcombe, -and •F.-T.—Turner-. —Methane Production
and Emission in a Texas Rice Field. Global Biofeochemical Cycles. 4(1), 47-
68, 1990.
31. Sass, R.L., F.M. Fisher, P.A. Harcombe, and F.T. Turner. Mitigation of
Methane Emissions From Rice Fields: Possible Adverse Effects of Incorporated
Rice Straw. Global Biogeochemical Cycles. 5(3), 275-287, 1991a.
32. Sass, R.L., F.M. Fisher, F.T. Turner, and M.F. Jund. Methane Emission From
Rice Fields as Influenced by Solar Radiation, Temperature, and Straw
Incorporation. Global Biogeochemical Cycles. 5(4), 335-350, 1991b.
33. Schutz, H. , A. Holzapfel-Pschom, R. Conrad, H. Rennenberg, and V. Seller.
A 3-Year Continuous Record on the Influence of Daytime, Season, and
Fertilizer Treatment on Methane Emission Rates from an Italian Rice Paddy.
J. of Geophvs. Res.. 94(D13), 16405-16416, 1989a.
.4-79
-------
34. Schutz, H., W. Seiler and H. Rennenberg. Soil and Land Use Related Sources
and Sinks of Methane (CH4) in the Context of the Global Methane Budget. Id:
Soils and the Greenhouse Effect. Ed. A.F. Bouwman. J. Wiley and Sons, N.Y.,
1990. 575 pp.
35. Seiler, W. , A. Holzapfel-Pschorn, R. Conrad, and D. Scharffe. Methane
Emission from Rice Paddies. J. of Atmos. Chen. 1, 241-268, 1984.
36. United Nations. FAO Production Yearbook, vols. 30, 31, 33, 35, 37, 39, 42,
43. Food and Agriculture Organization of the United Nations, Rome. 1977,
1978, 1980, 1982, 1984, 1986, 1989, 1990.
37. U.S. Dept. of Agriculture. Agricultural Statistics of People's Republic of
China, 1949-1982. International Economics Division, Economic Research
Service, Statistical Bulletin No. 714. U.S. Govt. Printing Office,
Washington, D.C., 1984.
38. Warneck, P. Chemistry of the Natural Atmosphere. Academic Press, N.Y., 1988.
753 pp.
39. Yagi, K. , and K. Minami. Effect of Organic Matter Application on Methane
Emission from Some Japanese Paddy Fields. Soli Sci. Plant Nutr.. 36(4), 599-
610, 1990.
4-80
-------
LIVESTOCK METHANE: SOURCES AND MANAGEMENT IMPACTS
by: Donald E. Johnson, T. Mark Hill and G.M. Ward
Metabolic Laboratory
Department of Animal Sciences
Colorado State University
Fort Collins, Colorado 80523
ABSTRACT
Herbivorous animals, particularly ruminants, have a digestive tract that
facilitates extensive symbiotic microbial digestion of dietary structural plant
carbohydrates. A byproduct of this symbiotic microbial process is an estimated
70 Tg of methane globally per year, primarily from cattle and buffalo. Cattle
methane emissions equal 6 ± 0.5% of their diet energy (2X by wt) for most global
conditions studied. Emissions by U.S. feedlot cattle are uniquely lower at about
3.52 of diet energy. A major lack of information on size, diet, class
distribution and percentage loss from developing country livestock precludes
accurate definition of this source, which is about 2/3 of global. Manure
disposal from livestock may produce an additional 12 Tg globally, primarily
through anaerobic lagoons. Possible ameliorative strategies include the
decreased use of lagoon disposal or the capture of this methane. General efforts
should concentrate on improving productivity of beef and dairy cattle production
systems, which will secondarily reduce methane.
4-81
-------
METHANE EMISSIONS: A BYPRODUCT OF ANIMAL MICROBIAL SYMBIOSIS
Digescive secretions by Che gastrointestinal tract of animals, per se, can
digest no structural components of plants. They can only digest the soluble
and/or starchy components. The only digestive enzymes that can unlock the
cellulose base of the structural components of plants are those produced by
microorganisms. Since about 75X of the photosynthetically fixed plant material
is cellulosic or structural, it perhaps is not surprising that many herbivorous
animals developed a symbiotic relationship with microorganisms in their
gastrointestinal tract to assist in utilizing these materials. All animals have
some microbial action in their gut, however, it is very extensive only in the
herbivores, particularly the ruminant herbivore. This fortuitous symbiotic
relationship between animal and microbe allows the utilization of vast tonnages
of cellulosic materials, i.e., grass, which would otherwise be left to decompose
on the earth's surface.
The microbes that function in the gastrointestinal tract, particularly the
carbohydrate utilizing anaerobes, require a sink to dispose of excess hydrogen
other than oxygen. Several species of the archebacteria fill this niche nicely.
These methanogens reduce C02 with the available hydrogen to produce methane. In
doing so, they doubly enhance the symbiotic relationship. First, they facilitate
the cellulolytic function of other bacteria and secondly, they increase the
supply of amino acids and vitamins to the host animal.
In general, the more extensive the gut microbial digestion of an animal
species, the higher the fraction of dietary loss as methane. Ruminant animals
(i.e., cattle, sheep and goats) with their large foregut fermentation vat, the
rumen, eructate or belch approximately 95X of the emissions from all animals.
Cattle in particular, because of their large numbers (1.2 billion), their large
size and appetites, coupled with this extensive symbiotic microbial fermentation
in their gastrointestinal tract account for some 71X of the approximately 70 Tg
of methane produced globally by animals each year. Another approximately 8X is
contributed by buffalo, with sheep and goats producing approximately 12X.
We have compiled the available observations of methane production (i.e.,
Table 1, cattle data) from the literature into a ruminant methane data base.
This data base includes 400 treatment mean observations of methane losses from
cattle and sheep, and minor numbers of measurements from other species. Methane
loss varied from 2.0 to 11.6X of dietary gross energy. Measurements Included
describe the many different weights and physiological states of the animals fed
and diets ranging from all forage to all concentrate diets or mixtures thereof.
An auxiliary spreadsheet lists approximately 1000 individual animal observations.
Many important concepts have emerged from our query and analysis of this
data set. The majority of the world's cattle, sheep and goats under normal
husbandry circumstances likely produce methane very close to 6X of their daily
diets gross energy (2X of the diet by weight). Although individual animals or
losses from specific dietary research circumstances can vary considerably, the
average for the vast majority of groups of ruminant livestock are likely to fall
between 5.5 to 6.5X. We must caution, however, that little experimental data is
available for two-thirds of the world's ruminants in developing countries.
Available evidence suggests similar percentage of emissions, but this supposition
needs confirmation. More importantly, data is skimpy or unavailable to describe
diet consumption, animal weight and class distribution.
4-82
-------
TABLE 1. SUMMARIZED VARIABLES FOR
RUMINANT METHANE
BEEF CATTLE
DATABASE
FROM
Variable
Minimum
Maximum
Mean
Observed methane, 2 gross energy
2.6
11.5
6.8
Observed methane, 2 digestible energy
3.3
16.7
9.8
Animal weight, kg
217
627
402
Dry matter intake, kg/d
2.1
10.9
5.4
Digestible energy, X
50.4
87.8
70.6
Level of intake, multiple of maintenance
0.4
2.9
1.5
One exception to this 62 rule is where cattle or sheep are fed very high
concentrate diets (> 802 grain and/or supplement). When fed these diets, likely
methane emissions will be 3.52 of gross energy. Frequently, they fall as low as
22. Such dietary circumstances occur almost exclusively in the U.S. feedlot
operations. Globally it has little reducing effect on emissions, approximately
27 million head of cattle fed for 140 days per year, with current emissions of
about .4 Tg/year.
Another important finding is the transitory effect of ionophores on
reduction of methane emissions. Ionophores are a class of antibiotic feed
additives which have been considered to suppress methane losses by 20-302. This
degree of suppression persists for some two weeks or less. Therefore, the
methane reduction effect of ionophores is more modest and primarily results from
a 6 to 72 reduced total feed requirements for production.
Another surprising finding was the uniqueness of one class of feedstuffs.
Brewery and distillery byproduct feeds produce about half as much methane as
other common feeds fed to ruminants (1). While of little impact globally because
of the limited amounts of such feed supplies, it could provide a clue to control
of methanogenesis.
An important principle influencing methane emissions from ruminant systems
is the inverse relationship between rate of productivity and methane losses,
especially when expressed per unit of animal product. Methane losses are closely
related to the amount of feed resource used to produce an animal product. An
increase in rate of production commonly decreases the feed/product by decreasing
the maintenance feed subsidy. Placing a beef calf directly into the feedlot in
the United States rather than the slower growth stocker phase preceding the
feedlot is expected to reduce the methane per lifetime of a steer by some 34Z
while producing the same amount of product (2). Perhaps more dramatically, the
supplementation of a moderate to low quality forage diet as might be employed in
Australia or South America, could increase the daily average gain from .35 kg up
to .7 kg. This increased rate of productivity would reduce the methane emissions
per lifetime of the steer from 170 to 100 kg, again without changing product.
Likewise, stimulating the rate of milk production by using bovine somatotropin
in the dairy cattle industry in the United States is expected to reduce methane
production by the industry some 92, essentially producing the same amount of milk
with less feed and less methane losses (3).
4-83
-------
One important additional source of methane indirectly emanating from the
livestock industry is that from manure disposal systems. The potential
production is huge, considerably larger than that coming directly from livestock,
however, measurements made in our laboratory (4) and in Australia (5) show a very
small production rate from manure disposed under simulated or actual range or
pasture situations. Thus, the major global disposition of manure on pasture
likely produces little methane. The critical question then becomes what fraction
of manure is disposed of by anaerobic lagoons a figure which is not known very
accurately. Our present best estimate of global manure methane adjusts the
disposal method data of Saffley et al. (6) to our estimates of range or pasture
production. With these suppositions, the estimate of global methane entry from
manure disposal approximates 12 Tg annually.
The work described in this paper was not funded by the U.S. Environmental Protection Agency. The
contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred.
REFERENCES
1. Wainman, F.W., Dewey, P.J., and Brewer, H.C. Feedingstuffs Evaluation
Unit, 4th Report, Rowett Research Institute, Aberdeen, 1984.
2. Johnson, D.E., Hill, T.K., and Ward, G.M. Methane emissions from cattle:
Global warming and management issues. Proc. Minn. Nutr. Conf. 1992 (in
press).
3. Johnson, D.E., Ward, G.M., and Torrent, J. The environmental impact of
bovine somatotropin use in dairy cattle. J. Envir. Oual. 21:157.
4. Lodman, D.W., Carmean, B.A., and Johnson, D.E. Estimates of methane
emissions from manure of U.S. Cattle. Chemos. (In press), 1992.
5. Williams, D.J. Methane production from manure from free-range cows.
Chemos. (in press) 1992.
6. Saffley, L.M., Jr., Casada, M.E., Woodbury, J.W., and Roos, K.F. Global
methane emissions from livestock and poultry manure. US-EPA Report, 1992.
4-84
-------
Paper 4-H
Ozone and Global Warming
by
Robert P. Hangebrauck, Air and Energy Engineering Research Laboratory
John W. Spence, Atmospheric Research ana Exposure Assessment Laboratory
ABSTRACT
Changes in several trace substances in the Earth's atmosphere are affecting global radiative forcing. Those
substances which seem to be in the greatest state of change now and projected into the future are carbon dioxide,
ozone (and its precursors and depieters), and aerosols. It is conceivable that countervailing changes in the
radiative forcing effects of these substances, especially ozone and aerosols, may be temporarily hiding or at least
changing the "greenhouse signal" — an unfortunate circumstance, particularly if the overall impacts that will
eventually occur are unpredictable or difficult to reverse quickly. If in fact the greenhouse signal is partially being
obscured at present, there is also potential for this effect becoming less significant in the decades ahead because
of 1) a continuation of increases in greenhouse gas emissions, 2) saturation of the troposphenc aerosol effect
plus controls on sulfur emissions, and 3) increasing troposphenc ozone. The substantial complexities in factors
affecting ozone and aerosols are discussed with emphasis on ozone and its precursors, including methane,
nonmethane hydrocarbons, carbon monoxide, and nitrogen oxides. Quantifying radiative forcing is of substantial
importance. EPA is undertaking research to enhance the ability to estimate indirect factors contributing to
forcing, including measures such Global Warming Potentials. Many of the important but difficult factors to resolve
are of the indirect type. A number of potential indirect forcing effects are identified along with an estimate of
direction (sign).
Introduction
Several trace substances affect global radiative forcing levels (changes in net downward flux of energy at the
tropopause). Some of the major ones are water vapor, carbon dioxide (C02), ozone (03), methane (CH4),
chlorofluorocarbons (CFCs), nitrous oxide (N,0), and aerosols. The continuing increase in the concentration of
these constituents has created considerable concern among scientists regarding the potential for climate change.
The observed increases are believed to have begun near the turn of the century with the continuing Industrial
Revolution and rapidly increasing technology. This period is marked with the beginning of an increasing energy
demand and conversion of forests to agricultural land to accommodate increasing world population. Radiative
forcing for individual trace gases and aerosols varies greatly, and depends on their concentrations. On the other
hand these constituents are controlled by various factors in addition to emissions, including mass transfer,
chemical interactions, and atmospheric lifetime. Ozone is a major factor in global radiative forcing but is not
well quantified because its high temporal and spatial variability make the quantification difficult. Ozone levels
in the troposphere depend on both transport of ozone from the stratosphere and on local chemistry. Production
of tropospheric ozone is dependent on concentrations of precursors (substances which produce ozone), including
CH4, carbon monoxide (CO), nonmethane hydrocarbons (NMHCs), and nitrogen oxides (NO,). Ozone levels in
the stratosphere now depend on concentrations of ozone-depleting substances, including CFCs, halons, other
halogenated organics, and N20. Temperature, amount of sunlight, global transport, amount of aerosols present,
and other factors are also of importance. On the other hand, methane and NMHCs inactivate 03-depleting
chlorine in the stratosphere. Radiative forcing for ozone is a strong function of altitude. Because of the
importance of ozone and its precursors as radiative and photochemical trace gases, EPA has accelerated its
research on global tropospheric ozone. EPA has had ongoing research on boundary-layer ozone for many years
in support of the ambient air standards dealing with the adverse effects of ozone on human health. Ozone is also
a toxicant to young trees and leafy crops (Reich, 1987; Heck, 1982).
Understanding of trace gas effects on global ozone and radiative forcing continues to emerge. For example, in
the stratosphere newly discovered heterogeneous reactions have been found to promote the formation of the
ozone hole in Antarctica and generally appear to promote the loss of ozone in the lower stratosphere. The loss
of ozone in the lower stratosphere has had an important impact on radiative forcing (Ramaswamy, 1992).
Similarly, initial attempts have been made recently to quantify the effects of tropospheric ozone on radiative
forcing (IPCC, 1990). This paper describes how azone fits into the global picture and covers some of EPA's related
EPA August, 1992
4-85
-------
research.
Obscuration of the Greenhouse Effect?
Global surface temperatures have risen some-
where near 0.3 to 0.6 *C over the last century
(IPCC, 1990). Figure 1 gives the global-mean
combined land-air and sea-surface tempera-
tures for the period 1861 -1989 based on histori-
cal records and various adjustments (IPCC,
1990). In addition to these past records, model
predictions of warming based on growth in green-
house gas loadings in the atmosphere imply
substantial future warming (IPCC, 1992). How-
ever, some - e.g., Michaels (1991) -- have expressed skepticism about the "popular vision" of global warming.
Example concerns include 1) adequacy of correction for the urban heat island effect, 2) less than expected
warming, relative to that predicted by modeling, and 3) the nature of how the warming has (or has not), taken
place -- such as, little increase in extreme high temperatures in the northern latitudes, However, it is likely that,
while the Earth has seen some global wanning, some potential warminghasbeen hidden by temporary offsetting
effects of other anthropogenic global changes. Changing ozone and aerosol levels over the last several decades
may have produced some net negative radiative forcing. This negative radiative forcing, along with a slight
decrease in solar irradiance, may have partially hidden the greenhouse signal predicted by general circulation
models (GCMs), especially in the Northern Hemisphere. Global changes other than warming may have resulted.
The changes taking place are difficult to predict as are any future consequences.
The effects of observed trends in atmospheric ozone on climate, while difficultto quantify, have been established
directionally (Lads, 1990; Ramaswamy, 1992). Radiative forcing from loss of ozone in the lower stratosphere
implies a cooling effect on the Earth's surface and in the stratosphere. Effects from an increase in tropospheric
ozone implies higher temperatures at the Earth's surface and in the free troposphere. More indirectly, increased
ozone in the atmospheric boundary layer reduces tree growth resulting in reduced carbon sequestration (Reich,
1987).
Estimates of the negative radiative forcing effects of aerosols range up to - 2 watts(W)/m2 for tropospheric sulfate
(Charlson, 1992). Penner (1991b) estimates from modeling work that aerosols from biomass burning could
contribute a radiative forcing mask of -1.8 W/m3 with aerosols from fossil fuel contributing even a larger negative
forcing. Kaufman (1991) concludes (with substantial uncertainties noted) that increased future sulfur dioxide
(S02) emissions from fossil fuel combustion will likely produce a cooling effect from increases in cloud
condensation nuclei. On the other hand biomass burning is seen overall to have a net warming effect, because
significant ozone precursors are injected into the atmosphere in addition to aerosols (Kaufman, 1991).
Stratospheric aerosols from the eruption of Mt. Pinatubo in 1991 should produce a very significant negative
forcing on the climate system (UNEPAVMO, 1991). Hansen (1992) estimates that the global mean climate
forcing due to Pinatubo will peak at about 4 W/m2 in early 1992, noting that this exceeds the accumulated forcing
due to all anthropogenic greenhouse gases added to the atmosphere since the industrial revolution. Global
climate changes that might be expected from increased tropospheric sulfate concentrations include the following
(Michaels, 1991):
• Increased cloudiness
• Enhanced brightening of low-level clouds near sulfate source regions
• A counteraction of daytime warming by the greenhouse forcing because of an increase in clouds
• Night warming from an increase in both clouds (especially stratocumulus) and greenhouse forcing
• A decrease in daily temperature range
" Decrease in ultraviolet (UV)-B in affected areas
• Concentration of these effects in the industrial Northern Hemisphere
The connection between aerosols and all of the above effects is by no means proven. Other factors must be
0.4
02
0.0
tr -0.2
-0.4
3 -0.6
1970
1990
1950
18B0
1910
¦ 370
tu
YEAfl
Figure t. Global-mean combined land-air and sea-surface
temperatures. 1961-1989, relative to the average for 1951-80.
(IPCC, 1990; Reproduced with permission.)
EPA August, 1992
-------
involved in reducing the predicted increases in the daily maximum temperature in a large portion of the Northern
Hemisphere (Karl, 1991). Some of these other factors could include changes in cloudiness and aerosol loadings,
but could also just be natural fluctuations of the climate system. These investigators were examining the nature
of and evidence for asymmetric diurnal temperature change in large areas of the Northern Hemisphere. These
are areas where, over the last several decades, most of the warming can be attributed to an increase of mean
minimum (mostly nighttime) temperatures with mean maximum (mostly daytime) temperatures displaying
little or no warming. These are also areas where there has been increasing extreme minimum temperatures but
little change in extreme maximum temperatures, leading to a decrease in the extreme temperature range.
Reduced Obscuration of the Greenhouse Effect?
Why should all this cause concern? Many climate scientists think that serious greenhouse effects are possible,
and that lack of a strong greenhouse signal may lead to inaction in dealing with emissions, particularly CO,
because of its long atmospheric lifetime. Furthermore, significant atmosphere-related global change may have
already taken place with unknown long-term consequences. In addition, if we are having substantial current
obscuration of the greenhouse signal, this obscuration may become less significant in the decades ahead because
of 1) a continuation of increases in greenhouse gas emissions, 2) saturation of the tropospheric aerosol effect, and
3) increasing tropospheric ozone.
Regarding saturation of the aerosol effect Kaufman
(1991) sees that, for a large increase in fossil fuel use,
emissions of S02 may saturate the cloud condensation
nuclei effect. For a doubling of C02, the aerosol radia-
tive forcing effect is estimated to be only 0.1 to 0.3 of
that for C02 compared to 0.4 to 8 times at current
emission levels.
Relative to tropospheric ozone, ozone is increasing on a
global basis with the rate of increase depending on
location and altitude. CH<( NMHCs, CO, and NOx are
all tropospheric ozone precursors, and their global
emissions are all increasing. In addition to surface-
based emission sources, emissions from aircraft at
cruise altitude seem to affect the ozone profile and are
similarly increasing (Wuebbles, 1990; Barrett, 1991).
How is O, Formed and Destroyed?
As the solar ultraviolet radiation enters the Earth's
atmosphere, the UV-C component is of sufficient energy to photodissociate oxygen (02), resulting in theformation
of ozone (03) in the upper stratosphere and UV-C absorption. UV-B radiation, on the other hand, penetrates to
lower altitudes and has enough energy to dissociate ozone (UV-B is absorbed in this process). Figure 2 illustrates
a typical ozone concentration profile for July at 60° N. This particular profile was derived from data which did
not include boundary layer measurements, and therefore does not show a typical increase near the Earth's
surface. The profile reveals a maximum mixing ratio near 30 km for this latitude. The ozone essentially filters
out most of the short wavelength ultraviolet radiation from the Sun. Figure 3 gives a simplified overall view of
the formation/destruction of ozone in the atmosphere. The destruction of ozone proceeds until these halogen
atoms are chemically bound in stable reservoir forms such as hydrochloric acid (HC1) and hydrobromic acid(HBr).
In general, atmospheric models utilizing homogeneous gas-phase chemistry provided a reasonable scientific tool
that explained the catalytic destruction of the ozone layer. With the formation of the ozone hole over the
Antarctic, atmospheric scientists began to investigate heterogeneous chemical mechanisms to explain the rapid
loss of ozone during the September-October time period in the Southern Hemisphere. The most critical aspect
of the heterogeneous mechanism is the destruction of active chlorine sinks by reactivating HC1 and chlorine
nitrate (ClONOj) and at the same time removing NOt from the stratosphere, Lower levels of NOt limit the degree
to which chlorine can be tied up as ClONO,. The sulfuric acid aerosols in the low to middle latitudes of the lower
WtHAUfcOA'Nk'
'ROFlLE F OR JULY
E
I
m
<
IOOO 7000 3000 *000 5000 0000 7000 »000
Volume Mixing Ratio (ppov)
toooo
Figure 2. Ozone aititudinal profile (data derived from chart
on p, 422 of Vol. II of Atmospheric Ozone 1985 (WMO,
1985).
EPA August, 1992
4-87
-------
stratosphere of the Northern and Southern Hemi-
spheres play a role similar to that of the polar
stratospheric ice crystals in contributing to ozone
depletion in the polar areas, especially Antarctica.
In the boundary layer ozone is formed photochemi-
cally from other gases (precursors). Within urban-
suburban areas, ozone formation is driven by the
emissions of the precursors NMHCs and NO,, in the
presence of sunlight. The formation of ozone in
remote areas of the boundary layer, in the free
troposphere, and in the lower stratosphere involves
the less reactive hydrocarbons such as methane.
The production of ozone is known to be complex and
is not linearly proportional to increases in precursor
emissions. Atmospheric modeling as well as photo-
chemical simulations indicate that the formation of
ozone may indeed be NCMimited over certain por-
tions of the globe. These studies indicated that
How is 03 formed and destroyed?
Formation:
02 ~ UV
02 ~O-
Deslrudon:
->0 + 0
-»Q3
CI + 03 —> CtO ~ 02
co ~ o -» a ~ 02
(Nat) O + 03 —» 2 02
Stratosphere
coNosj,
HQ
r ci
CD
^ HN03
Troposphere
Formation (high NOx):
(Nat reaction)
CH4 ~ 402 —> CH20 ~ H20 ~ 203
Cp*8QB->CQgtW
Oesirucrcn (NO* < 20 pptv):
(Nat reaction)
CH4 ~ 02 -» CH20 ~ H20
CO ~ Q3 —> CQ2 ~ 02
Boundary Layer
Figure 3. Overview of ozone formation/destruction in the
atmosphere.
concentrations ofNOi above a range of 10 to 25 pptv are required for ozone production, while lower concentrations
would lead to destruction (Hameed, 1979; Lin,
1988). This implies that there are areas within the
boundary layer as well as free troposphere where
the photochemistry is forming ozone and other
areas where photochemistry is destroying ozone.
Scientists at Lawrence Liverroore National Labo-
ratory (LLNL) are developing a 3-D chemistry
model to predict global atmospheric loadings and
* distributions of NO, including nitric acid from
i anthropogenic and biogenic sources (Penner, 1991a;
Dignon, 1992). A global map of spatially allocated
NO, emissions in Figure 4 illustrates a typical
input used in the model. The model also predicts
global wet and dry deposition of nitrogen species.
Figure 4. Spatially allocated global NO, emissions from Developing the capability to model NO from its
anthropogenic and natural sources in 1980.
5f
5I
15
DO
01
oafsa*
006 a
sources is the first step in developing a 3-D Global Chemical
Model for predicting atmospheric loadings of ozone and
other greenhouse gases from emissions. Ozone is also
destroyed by surfaces, by reactions with olefmic gases, and
by photodissociation that leads to production of hydroxy!
(OH) radicals. The oxidizing power of the Earth's atmo-
sphere is controlled by the abundance of these radicals. It is
estimated that OH radicals account for about 85% of the loss
of methane (Cicerone, 1988). Some of the directional effects
of precursor emissions on ozone-related atmospheric chem-
istry are indicated in Figure 5 as derived from UNEP/WMO
(1991). For example, increasing the precursor CH4 will
decrease OH, increase 03, and increase the lifetime of
various organics such as HCFCs.
Trends in Ozone
Figure 6 illustrates the change in total column ozone in the period 1979 - 1991 from data reported by UNEP/
Increased
emission
OH
03
Ufellme (CH4,
HCFC, HFC)
CH4
-
~
~
NOx
~
~
•
CO
-
*
~
NMHC
-
~
~
Figure S. Precursor influences on atmospheric
chemistry (derived from UNEP/WMO, 1991).
EPA August, 1992
4-88
-------
Data from UNCwwMp 1M1 Oloiw
AlMMMMl
~ *y<
Q
¦ 45" k
DvoMii S«p-Nov
Figure 6. 1979-1991 total ozone column change
(TOMS satellite data.)
WMO (1991) for the Total Ozone Mapping Spectrometer
(TOMS) satellite. While this show substantial depletions,
more recent data reported by NASA (1992) show much greater
depletions, including in the tropics. In addition, measure-
ments show record concentrations of CIO of 1.5 ppbv over
northern New England in January 1992 as part of the Air-
borne Arctic Stratospheric Expedition (NASA, 1992). Abun-
dance of chlorine monoxide (CIO) in the lower stratosphere at
northern middle-latitudes is greater than predicted by mod-
els containing only gas-phase chemistry (UNEP/WMO, 1991).
36.0
34.0
32.0
30.0
28.0
26.0
24.0
22.0
20.0
18.0
16.0
14.0
12.0
100
BO
6.0
4 0
2.0
1
/
r
\ (
: \
\
V
N
V
N
V
-1.0
<0.5
0.0
0.6
1.0
Ay wigs a Ozsna.
Figure 7. Ozone aItitudinaI trend estimates tor the
Northern Hemisphere (1970 - 1986).
Information on the vertical atmospheric distribution of ozone can be derived from ozonesonde, Umkehr,
Stratospheric Aerosol and Gas Experiment (SAGE), and Upper Atmosphere Research Satellite (UARS) data.
Unfortunately, most of the long-term measurements, particularly for the ozonesondes, are within the northern
middle-latitudes. This hinders the development of reliable global vertical distributions for the Southern
Hemisphere. There are concerns regarding calibration techniques for the long-term data from many of the
northern ozonesonde stations. Nevertheless, the average percent per year change of ozone from 1970 to 1986
as a function of altitude for nine northern stations is shown in Figure 7 (UNEP/WMO, 1989). This historical
ozonesonde record reveals thatozone appears to be increasing within the free troposphere and decreasing within
the lower stratosphere. However, variations at individual stations show widely varying profile changes for
different latitudes. Tropospheric ozone trends derived from ozonesonde data have accuracy limitations, but
could be improved by expansion of monitoring capability (Prinn, 1988). Homogeneous chemistry 1-D models
predict the ozone increases within the free troposphere but fail to predict the decreased ozone observed within
the lower stratosphere. The change in the slope of the ozone profile that is observed in the troposphere below the
tropopause may be due to the injection of ozone precursors by aircraft (Wuebbles, 1990; Kinnison, 1991).
Tropospheric ozone trends continue to be measured at ozonesonde stations; however, measurements in the upper
troposphere are sparse, and this is where ozone has its greatest radiative forcing effect.
Satellite data from the TOMS and the SAGE have been recently used to derive global maps of ozone within the
troposphere and boundary layer (Fishman, 1991). Tropospheric ozone is derived as the residual or difference
between the coincident TOMS and SAGE measurements between 50° S and 50° N. Hie residual, which is a
relatively small difference between two larger values, represents the ozone column in Dobson units (ozone
molecules per cm2) within the troposphere and boundary layer. Over a 10-year period the averaged seasonal
depictions show the residual ozone to begin its formation in the Northern Hemisphere during March-May,
reaching a maximum during June-August. In the Southern Hemisphere, ozone forms during the September-
November season. What is so surprising is the size of the ozone plumes. The entire Northern Hemisphere is
engulfed in an ozone plume that spans the Atlantic and much of the Pacific Ocean for the June-August season.
EPA August, 1992
-------
Figure 8. Satellite-derived troposphere ozone residuals
centered off the west coast o< Africa.
In the Southern Hemisphere the ozone plume forms across lower Africa and tails across to Australia. It spans
the Atlantic Ocean to South America. The formation of this plume appears to be associated with the biomass
burning that occurs annually during the spring in the Southern Hemisphere. The formation of the observed
ozone in the Northern Hemisphere is consistent with the photochemical production of ozone from its precursors
during the spring and summer months.
Atmospheric scientists at NASA Langley Research Center are using the TOMS and SAGE data to derive daily
residual maps for global tropospheric ozone for EPA. Since SAGE provides less than 5000 measurements per
year, daily measurements are linearly interpolated from 60 day averages. The daily residual is obtained by
subtracting the interpolated SAGE daily measurements from 3-day averages of the TOMS measurements and
has a resolution of 2.5° latitude by 5.0 D longitude. Computerized images ofthe daily residual from 1985 to 1990
are prepared by the EPA's Scientific Visualization Laboratory at RTP. A 4-day sequence for September-October
1988 is shown in Figure 8 thatis believed to represent the formation of tropospheric ozone from the biomass burn
in the springtime in the Southern Hemisphere. Scientists at NASA Langley Research Center are using the video
computerized images to assist in the planning of the TRACER A Monitoring Program of the Biomass Burn in
1992. Scientists at EPA are comparing the satellite-derived daily residual 03 data for the Northern Hemisphere
with ozone ground-based (AIRES), ozonesonde, and meteorological measurements. If the comparative analysis
shows promising results, the satellite-derived ozone residuals will provide insight into the formation, transport,
and fate of global tropospheric ozone, and may serve as a useful tool in the development of global chemical models
that predict ozone concentrations from precursor emissions. EPA and NASA scientists are planning to develop
another sequence of daily residuals using measurements from the Solar Backscatter Ultraviolet Spectrometer
(SBUV) that flies on the Nimbus 7 satellite with the TOMS. Comparative analysis ofthe two daily residual data
bases should provide insight into the techniques for deriving global tropospheric ozone from satellite data.
EPA August, 1992
4-90
-------
Trends in Aerosols/Suifates
It has been concluded that fossil fuel emissions over the past century have increased the tropospheric sulfate
aerosol concentrations (UNEP/WMO, 1991). World total sulfur emissions are estimated to be 147 TgS with about
80 TgS coming mainly from fossil fuel combustion (IPCC, 1990). Husar (1989) shows a general correlation
between sulfur emissions and extinction coefficient which is a function of visibility. Mayewski (1990) has
determined, based on ice core data, that the anthropogenic sulfate loadings in remote areas of the Northern
Hemisphere are now as high as or higher than the maximum loading from many past volcanic eruptions.
Carbonyl sulfide (COS) is the most abundant sulfur-containing trace gas in the remote atmosphere and, with a
lifetime of 2 to 6 years, is, via photolysis, a major source of aerosol sulfate in the stratosphere along with volcanic
eruptions (UNEP/WMO, 1991). Sources of COS include anthropogenic activities, soils, biomass burning, and
oxidation of carbon disulfide (UNEP/WMO, 1991). Long-term observational records show a 40-50% increase in
stratospheric sulfate aerosols (UNEP/WMO, 1991).
Trends in Ozone Precursors
Source emissions of CH4, NOt, CO, and NMHCs
have all been increasing. Figure 9 shows the
IPCC predictions of future ozone precursor emis-
sions. Atmospheric concentrations/distributions
have been quantified for CH4 and CO, but are
harder to do for the more reactive species, N0I
and NMHCs.
The atmospheric concentration of CH4 has been
increasing at a rate of about 1% per year up to the
last few years. Currently this growth rate ap-
pears to have fallen sharply with no completely
satisfactory explanation. However, there is a
possibility that OH is increasing at 1.0±0.8% per
year (UNEP/WMO, 1991). The most recent data
for methane, as taken from Khalil(1990), are given in Figure 10. Unfortunately, the data after 1988 are still not
available. The data are undergoing quality assurance review, but should be available this summer in the journal
Chemosphere and a new book entitled The Global
Methane Cycle: Its Sources, Sinks, Distributions and
Role in Global Climate Change. Figure 11 represents 5
years of methane measurements from remote marine
sites within the Geophysical Climate Change Sampling
Network. The averaged atmospheric concentration of
IPCC Trace Gas Projections!
NMHC
N IHC
Figure 9. Projected ozone precursor emissions.
,1700
1680
1660"
1640
v 1620
1600
1SS0
1 560
1 540
1 980
1 9B2
YEAR
r i i r
1984 1966 1988
Figure 10. Atmospheric methane concentration.
Figure 11. Hemispheric methane cycles with time.
(IPCC, 1990; Reproduced with permission.)
EPA August, 1992
4-91
-------
methane is about 1.76 ppmv in the Northern Hemisphere and about 1.68 ppmv in the Southern Hemisphere. The
seasonality is observed to vary with latitude and is repeatable over time in the hemispheres CIPCC, 1990). Note
the increase in methane concentration right up to the arctic polar area. The primary emission sources of methane
include: tundra/bogs/swamps, rice cultivation, biomass burning, livestock, coal mining, oil, and natural gas
systems, landfills, and a variety of industrial processes (such as coke and petrochemicals production). Methane
isotopic studies suggest that 20% of global methane is from fossil fuel and 10% is from biomass burning (UNEP/
WMO, 1991).
The atmospheric concentration of carbon monoxide is increasing at about 1% per year in the Northern
Hemisphere but is not increasing perceptibly in the Southern Hemisphere. The concentration is about 120 ppbv
in the Northern Hemisphere and about 60 ppbv in the Southern Hemisphere for a hemispheric ratio of about 2
(Wuebbles, 1991). The primary sources of atmospheric CO are atmospheric oxidation of methane and
nonmethane hydrocarbons, biomass combustion-related sources such as forest clearing, and fossil-fuel combus-
tion (primarily transportation related).
Measurements of NMHCs are not adequate to establish trends
with the exception of ethane (UNEP/WMO, 1991). Ehhalt
(1991) has provided evidence of a trend for ethane over the
Northern Hemisphere of 0.9±0.3% per year. Perhaps the best
indicator of trend for NMHCs is the projected future emissions
(see Figure 9). The primary sources of atmospheric NMHCs are
biogenic in nature, including trees, oceans, and grasslands
(UNEPAVMO, 1991). Anthropogenic sources include gasoline
use, other petroleum-based solvents/chemicals, fuel wood use,
biomass burning, and waste disposal (Watson, 1991). The
reactivity and general variability of NMHCs in the atmosphere
make it difficult to establish trends. NMHCs and their reaction
by-products are found throughout the atmosphere. For ex-
ample, the work of Greenberg (1990) indicates that many
NMHCs may exist throughout the troposphere. Longer photo-
chemical lifetimes, attributed to lower temperatures at high
altitudes and latitudes, suggest to the investigators "that
NMHCs may be present throughout the troposphere in many
regions of the Northern Hemisphere." The more reactive NMHCs are generally found in smaller quantities
relative to the less reactive compounds. Transport mechanisms such as cumulus clouds, dry convection, cold
fronts, and flow of air over mountains are suggested as means for the substantial altitudinal dispersal in spite
of atmospheric lifetime considerations. Figure 12 illustrates some of the data reported by Greenberg — in this
case for total NMHCs at 66° N latitude. As can be seen, substantial NMHC mixing ratios are found high in the
troposphere.
Penner(1991a) and Dignon (1992) have estimated sources and distributions globally of natural and anthropo-
genic emissions of NO,. Northern Hemisphere anthropogenic sources are seen to be mainly fossil fuel
combustion, and Southern Hemisphere anthropogenic contributions are seen to be a combination of fossil fuel
emissions and biomass burning. Fossil fuel combustion, which consists of both stationary and mobile sources,
is believed to account for about 50% of the estimated total emissions of global NOt. Ifbiomass burning is included,
the total anthropogenic contributions become more like 75%. However, there is considerably more uncertainty
in the emissions from biomass burning than from fossil fuel combustion in stationary sources. The NASA project
(TRACER A) to characterize the 1992 biomass burn in the Southern Hemisphere should provide better
quantification of the emissions. Natural sources include lightning, soil microbial activity, and input into the free
troposphere from the photodissociation of N20 in the stratosphere. Large uncertainties are also associated with
the emissions from the natural sources. Sufficient atmospheric measurements of NO, are not available to project
global concentration trends. The best indication of NOf concentration trends isfromnitrate (N03) measurements
taken from ice cores. Analyses for N03 in ice cores from Greenland and Switzerland show large increases in NO,
since the turn of the century (Neftel, 1985; Wagenbach, 1988).
Lao! Id* i 6 "N
O 10 20 30 40 50 60 70 80 90
NHMC Mixing Ratio, pptv
Figure 12. A total NMHCs profile for 66 0 N
latitude from Greenberg (1990).
EPA August, 1992
4-92
-------
Radiative Forcing
Determining the radiative forcing caused by various trace substances is obviously difficult, with major questions
still to be answered. Multiple effects for the same substance are possible in different parts of the atmosphere.
Many of the complex questions arise from the indirect effects of trace substances. The recent IPCC (1990) report
provided initial quantification of some of the indirect effects in terms of Global Warming Potentials (GWPs),
especially for ozone and its precursors. But in later UNEPAVMO updates of the IPCC report, reservations have
been expressed about the accuracy of the initial indirect GWPs. Also, the rather large negative-forcing indirect
effect was identified for CFC-related ozone depletion in the lower stratosphere . GWPs are evolving with
refinements taking into account such factors as sensitivity of the GWPs to the assumed background atmosphere
(Wuebbles, 1992). Many of the indirect radiative forcing effects, for which indirect GWPs would be generated,
require modeling to provide quantification, but for the most part the tropospheric models necessary to do the
evaluation have not been fully developed and validated. EPA is collaborating with the Lawrence Livermore
National Laboratory to identify, quantify, and refine potential indirect forcing effects for the various trace
substances, especially as they relate to tropospheric ozone and its precursors.
As a greenhouse gas, the radiative properties or forcing for a given quantity of ozone is a function of altitude
{Lacis, 1990). The radiative forcing of ozone has little affect within the boundary layer but maximizes within the
free troposphere at the tropopause. In the lower stratosphere, the forcing is positive but diminishes with altitude.
It is also a function of latitude. Hansen (1991) recently reported (at the 1991 AGU meeting in San Francisco)
the impacts of atmospheric ozone on radiative forcing using the GISS GCM . When ozone between 10-20 km
within the lower stratosphere was removed, the upper troposphere was found to cool by several degrees Celsius
and the Earth's surface temperature cooled by 1-2° C. This simulation provides insight into the radiative effects
of ozone removal that is believed to be now occurring by heterogeneous reactions in the lower stratosphere. When
ozone above 35 km was removed in the model simulations, the stratosphere cooled, and a warming of about 2 °
C at the surface was observed. Hie warming of the surface in this case can be associated with an increase in
radiant energy transmitted through the atmosphere. Increasing ozone in the free troposphere by a factor of 10
resulted in surface warming as a function of latitude. While the GISS GCM simulations point out the radiative
importance of atmospheric ozone, the simulations also indicate a need for better ozone measurements in the
troposphere and stratosphere to establish ozone profile trends. EPA scientists are also initiating studies to
parameterize atmospheric transport processes between the boundary layer and free troposphere, such as cloud
venting, for incorporation into global chemistry models. These processes inject ozone and its precursors into the
troposphere. The atmospheric lifetime of ozone increases significantly in the troposphere where the temperature
decreases with altitude up to the tropopause. These studies are intended to enhance model determination of
global radiative influences due to changes in tropospheric ozone.
Table 1 illustrates some of the direct and indirect effects related to radiative forcing and their likely directional
effects. This table is not provided as an accurate, comprehensive compilation of all potential important effects,
but does attempt to show that there are many effects. Almost all the effects from the inorganic gases covered
in Table 1 are directionally positive. Water vapor has a positive forcing effect in both the troposphere and
stratosphere; however, there are also feedback effects related to water including clouds, snow cover, sea ice cover
and precipitation. Hydrogen is also increasing in the atmosphere (Khalil, 1990) and can contribute water vapor
to the stratosphere causing an indirect positive effect. CO has positive indirect effects associated with 1)
production of ozone in the troposphere and 2) additional C03 as the final product of oxidation. Nitrous oxide is
a direct acting greenhouse gas and might also cause positive forcing via its depletion of upper stratospheric ozone.
NO] have a positive indirect forcing via production of tropospheric ozone. Radiative forcing of tropospheric ozone
may be especially sensitive to NO, emitted by aircraft at cruise altitude (Wuebbles, 1990; Kinnison, 1991;
Johnson, 1992). NO, may also have anegative forcing effect in the stratosphere via stratospheric ozone depletion.
All of the effects identified for methane in Table 1 are positive. It has a direct effect, and it produces positive
effects from ozone production in both the troposphere and the stratosphere. It has a positive effect via
deactivation of active stratospheric halogens which destroy stratospheric ozone. It produces stratospheric water
vapor, and it ultimately ends up as C02, a greenhouse gas. All of the effects listed for NMHC s in Table 1 are also
positive. NMHCs produce positive forcing from ozone formation in the troposphere and stratosphere. Longer-
lived NMHCs that make it to the stratosphere would deactivate active CI, and Br,, thus reducing stratospheric
ozone depletion. NMHCs, like CO and methane, will also eventually oxidize to the greenhouse gas, C02.
EPA August, 1992
-------
Table 1. Partial listing of direct and indirect
effects related to radiative forcing.
BIO
Bydrof*
COS
CO
NSO
NO*
CH4 (aft ~>
H20rTmp. (direct ~}
H2CV5lni. (direct ~)
H^StreL H30 (indirect ~ >
C02 (direct ~}
CC?2/Str*L 03 ir.creeee vu iirei. cvohnf (direct ~)
COTrop 03 indirect Vj
CCVC02 (indirect *)
N20 (direct*)
V2(V%nt G3 depletion (indirect *)
NOi lircnA/Upptr tnp. 03 (indirect ~ )
NOi/Stnl. 03 reduction (indirect -)
Oifuia
Halogenated organics would appear to cause both
positive and negative forcing. CFCs such as CFC-11
and CFC-12 are strong absorbers causing direct green-
house forcing in the troposphere. On the other hand
such long-lived halogenated organics will cause strato-
spheric ozone depletion, thereby causing negative forc-
ing. Aerosols constitute another set of forcing effects
that are not well quantified and complex, but it would
appear that their effects are mostly negative. Aero-
sols, such as sulfuric acid, can cause negative forcing
through reflective scattering of solar radiation. They
can also enhance cloud formation and reflectivity in
the troposphere. Aerosols containing carbon black, on
the other hand, may induce positive forcing via ab-
sorption of radiant energy. Finally, sulfuric acid
aerosols in the lower stratosphere probably enhance
ozone depletion via heterogeneous catalysis and de-
struction of sink species.
Conclusions
Changes in several trace substances in the Earth's
atmosphere are affecting global radiativeforcing. Those
substances which seem have the largest changes oc-
curring now and projected into the future are C0J(
ozone (and its precursors and depleters), and aerosols.
It is conceivable that countervailing changes in the
radiativeforcing effects of these substances, especially
ozone and aerosols, may be temporarily hiding or at
least changing the "greenhouse signal." If in fact the
greenhouse signal is being partially obscured at
present, there is also potential for this effect's becom-
ing less significant in the decades ahead because of 1)
a continuation of increases in greenhouse gas emis-
sions, 2) saturation of the tropospheric aerosol effect
plus controls on S02 emissions, and 3) increasing
tropospheric ozone. Recent findings by international
scientists working toward more accurate assessment
of future forcing make it clear that indirect effects are playing an important role and are poorly quantified at
present. The substantial complexities in factors affecting ozone and aerosols are discussed with emphasis on
ozone and its precursors, including methane, NMHCs, CO, and N0X. A substantial number of potential forcing
effects are identified along with an estimate of direction (sign). Quantifying radiative forcing and its sources are
of substantial importance for future prevention and mitigation efforts, and to this end EPAis helping with efforts
to enhance the ability to estimate both the direct and indirect factors contributing to forcing. Emphasis is being
placed on better estimates of current and projected emissions and measures of radiative forcing such as Global
Warming Potentials (GWPs). Research efforts are focused on enhanced emission/mitigation data and projec-
tions, improved data on global atmospheric trends, and enhanced capabilities for theoretical prediction via
improved models.
CH4 (direct ~)
CHVTrop. 03(tnd>rect ~)
CHVSlret. 03 (indirect ~)
CHVStrat. 03 increaae via CU/Ent deactivation (indirect ~)
CHVSini. H20 (indirect * I
CH4C02 (indirect ~!
NMKCelalte)
NMHC*Trop, 03 (indirect *)
Ethane, prupane, etc.
NMHCaRtrei 03 (indirect t|
Ethane, propane, etc.
NMHCafttrvL 03 inc. via CU/Brv deactivation (indirect »>
Ethane, propane, etc.
NMHCVC02 (indirect ~)
Ualtfiuud Orfmnica (+ & ->
CFC-11 (djract ~>
CFC-12 (direct ~)
CFC-113 (direct*)
CFC-22 (direct «¦)
CUABrt/Lomn Stmt 03 depletion (indirect «>
CUABn/Upper Strau 03 depletion (indirect ~)
Ketone (direct •)
CH3Br (direct ~)
AatMiIa (Mostly -)
AeroioWStrauTrop. (direct •)
Aareaaia'Trop (indirect •)
Aereaait (Carbon bleckVTivp. (direct
Aan>aoJa
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication.
EPA August, 1992
4-94
-------
References
Barrett, M., Aircraft Pollution: Environmental Impacts and Future Solutions, WWF-World Wide Fund for
Nature Research Paper, Switzerland (1991).
Charlson, R. J., Schwrtz, S. E., Hales, J. M., Cess, J. A., Coakley, J. A., Jr., Hansen, J. E., and Hofman, D. J.,
Climate Forcing by Anthropogenic Aerosols, Science, 255, 423-430 (1992).
Cicerone, R. J,, and Oremland, R. S., Biogeochemical Aspects of Atmospheric Methane, Global Biogeochem.
Cycles, 2, 299-327(1988).
Dignon, J., Penner, J. E., Atherton, C. S., and Walton, J. J., Atmospheric Reactive Nitrogen: A Model Study of
Natural and Anthropogenic Sources and the Role of Microbial Soil Emissions, Proceedings of CHEMRAWN VII
World Conf. on the Chemistry of the Atmosphere: Its Impact on Global Change, Baltimore, MD (1992).
Ehhalt, D., Schmidt, U., Zander, R., Demoulin, P., and Rinsland, C., Seasonal Cycle and Secular Trend of the
Total and Tropospheric Column Abundance of Ethane Above the Jungfraujoch, J. Geophys. Res, 96, 4987-
4994(1991).
Fishman, J., Probing the Planetary Pollution from Space, ES&T, 25, 613-621 (1991).
Greenberg, J. P., Zimmerman, P. R., and Haagenson, P., Tropospheric Hydrocarbon and CO Profiles, Over the
U.S. West Coast and Alaska, Journal of Geophysical Research, 95,14015-14026 (1990).
Hameed, S., Pinto, J. P., and Stewart, R. W., Sensitivity of the Predicted CO-OH-CH4 Perturbation to
Tropospheric NO, Concentrations, JGR 84(C2): 763-768 (1979).
Hansen, Can the Climate be Engineered?, AGU Fall Meeting Invited Talk, San Francisco (January 1991).
Hansen, J., Lacis, A., Ruedy, R., and Sato M., Potential climate impact of Mount Pinatubo eruption, Geophysical
Research Letters, 19, 215-218 (1992).
Heck, W. W., Taylor, O. C., Adams, R., Bingham, G., Miller, J., Preston, E., and Weinstein, L., Assessment of Crop
Loss from Ozone, JAPCA, 32:353-361 (1982).
Husar, R. B., and Wilson, W. E., Trends of Seasonal Haziness and Sulfur Emissions Over the Eastern U.S.,
Transactions: Visibility and Fine Particles, An AWMA/EPA Specialty Conference, Air and Waste Management
Association (1989).
Intergovernmental Panel on Climate Change (IPCC), Climate Change, The IPCC Scientific Assessment, World
Meterological Organization/United Nations Environment Programme (1990).
IPCC, 1992 IPCC Supplement, Intergovernmental Panel on Climate Change, UNEP/WMO (1992).
Johnson, C., Henshaw, J., and Mclnnes, G., Impact of Aircraft and Surface Emissions of Nitrogen Oxides on
Tropospheric Ozone and Global Warming, Nature, 355, 69-71 (1992).
Karl, T. R., Kukla, G., Razuvayev, N., Changery, M. J., Quayle, R. G., Heim, R. R., Easterling, D. R., and Fu, C.
B., Global Warming: Evidence for Asymmetric Diurnal Temperature Change, Geophysical Research Letters, 18,
12, 2253-2256 (December 1991).
Kaufman, Y. J., Fraser, R. S., and Mahoney, R. L., Fossil Fuel and Biomass Burning Effect on Climate - Heating
or Cooling?, American Meteorological Society, 4, 578-588 (1991).
Khalil, M. A. K., and Rasmussen, R. A., Atmospheric Methane: Recent Global Trends, Environmental Science
EPA August, 1992
4-95
-------
and Technology, 24, 549-553 (1990).
Kinnison, D. E., and Wuebbles, D. J., Future Aircraft and Potential Effects on Stratospheric Ozone and Climate,
Proceedings of the 42nd Congress of the International Astronautical Federation, Montreal, Canada (1991).
Lacis, A. A., Wuebbles, D. J., and Logan, J. A., Radiative Forcing of Climate Changes in the Vertical Distribution
of Ozone, Journal of Geophysical Research, 95, 9971-9981 (1990).
Lin, X., Trainer, M., and Liu, S. C., On the nonlinearity of the tropospheric ozone production, JGR, 93,15879-
15888 (1988).
Mayewski, P. A., Lyons, W. B., Spencer, M. J., Twickler, M. S., Bock, C. F., and Whitlow, S., An Ice-Core Record
of Atmospheric Response to Anthropogenic Sulphate and Nitrate," Nature, 348, 22 (1990).
Michaels, P. J., Journal of Coal Quality, 10, 1 (March 1991).
NASA, Summary Statement: Second Airborne Arctic Stratospheric Expedition (February 3, 1992).
Neftel, A., Beer, J., Oeschger, H., Zurcher, F., and Rinkel, R. C., Sulfate and Nitrate Concentrations in Snow from
South Greenland, Nature, 314, 611-613 (1985).
Penner, J. E., Atherton, C. S., Dignon, J., Ghan, S. J., and Walton, J. J., Tropospheric Nitrogen: A Three:
Dimensional Study of Sources, Distributions, and Deposition, J. Geophys. Res., 96, 959-990 (1991a).
Penner, J. E., Ghan, S. J., and Walton, J. J., The Role of Biomass Burning in the Budget and Cycle of
Carbonaceous Soot Aerosols and Their Climate Impact, Global Biomass Burning: Atmospheric, Climatic, and
Biospheric Implications, J. S. Levine, editor, MIT Press, Cambridge, MA (199 lb).
Prinn, R. G., Toward an Improved Global Network for Determination of Tropospheric Ozone Climatology and
Trends, J. Atmospheric Chemistry, 6, 281-98 (1988).
Ramaswamy, V., Schwarzkopf, M. D., and Shine, K. P., Radiative Forcing of Climate from Halocarbon-Induced
Global Stratospheric Ozone Loss, Nature, 355, 810-812 (1992).
Reich, P. B., Quantifying Plant Response to Ozone: a Unifying Theory, Tree Physiology, 3, 63-91 (1987).
UNEP/WMO , Scientific Assessment of Stratospheric Ozone Depletion: 1991 (1991).
UNEP/WMO , Scientific Assessment of Stratospheric Ozone, (1989).
Wagenbach, D., Munnich, K., Schotterer, U., and Oeschger, H., Annals of Glaciology, 10, 182-187 (1988).
Watson, J. J., Probert, J. A., and Piccot, S. D., Global Inventory of Volatile Organic Compound Emissions from
Anthropogenic Sources, EPA-600/8-91-002 (NTIS PB91-161687) (January 1991).
WMO, Atmospheric Ozone 1985, Volume II, Report No. 16 (1985).
Wuebbles, D. J., and Patten, K O., Sensitivity of Global Warming Potentials to the Assumed Background
Atmosphere, Lawrence Livermore National Laboratory, UCRL-ID-109847 (1992).
Wuebbles, D. J. and Edmonds, J., Primer on Greenhouse Gases, Lewis Publishers (1991).
Wuebbles, D. J., and Kinnison, D. E., Sensitivity of Stratospheric Ozone to Present and Possible Future Aircraft
Emissions, Air Traffic and the Environment, U. Schumann, editor, Springs-Verlag (1990).
EPA August, 1992
4-96
-------
Overview of Methane Energy and Environmental Research Programme
in the United Kingdom
Suzanne A. Evans
Project Officer, Landfill Microbiology
Anton van San ten.
Section Leader, Waste Combustion
Paul S. Maryan
Section Leader, Agriculture and Forestry
Caroline A. Foster
Project Officer, Arable Energy Forestry
Keith M. Richards
Programme Manager. Biofuels
Energy Technology Support Unit,
Harwell Laboratory,
Oxon OX11 ORA
United Kingdom
ABSTRACT
The biofuels programme of research and development forms an important part of the
UK Department of Energy's renewable energy programme.
This programme began in the mid 1970's as a response to the oil crisis of that time
and was part of a much wider look at alternative energy supplies. The initial
driving force for Biofuels and other renewables research was the prospect of greater
diversity, and hence security of energy supply for the nation.
More recently, concerns over the environment and the need for 'sustainable' sources
of energy, has added further strength to the case for using 'environmentally friendly'
renewables.
This paper reviews the history of the biofuels programme, its present content and
considers where future emphasis might lie.
The work described in this paper was not funded by the U.S. Environmental Protection Agency. The
contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred.
©Crown Copyright 1992
(Reproduced with permission.)
4-97
-------
INTRODUCTION
Biofuels: Status
Biofuels are any solid, liquid or gaseous fuels produced from organic materials
which are derived either directly from plants or indirectly from industrial, commercial,
domestic or agricultural wastes. They are amongst the most economically attractive
sources of renewable energy offering significant potential and also environmental
benefits locally, nationally and globally.
Estimates * suggest that some 14% of the world's energy (around 55EJ) is
satisfied by the use of biofuels, making it by far the most important of the renewable
sources. Renewable energy contribution in total is of the order of 20%. (See figure 1).
The most important use of biofuels is made by 'third world' or developing
countries where use in rural areas often exceeds 90% of total energy requirements. In
'developed' countries use, as a proportion of total energy need, is much less, around
2-2.5% on average in the EEC and 4% in the USA. A recent survey of IEA nations
participating in the Bioenergy Agreement observed that the highest level of contribution
was no more than 15%. (See figure 2).
In the United Kingdom, use of biofuels amounted to a quarter of one percent of
total primary energy needs in 1990, an estimated 23% of the total renewable energy
supply. (See figure 3). Some of the reasons for the limited use of biofuels in the UK
are:
The UK is a fossil fuel rich nation and traditional sources of energy including
coal, oil, natural gas (and also nuclear) are well established in the market place.
Biofuels are very diverse in nature, disaggregated and (historically) satisfy only
limited, local 'niche' markets.
Biofuels are often bulky and less convenient to handle than traditional fuels
Biofuels have a 'rustic' image and hence are often dismissed by large energy
producers/users.
Biofuels often appear uneconomic at the medium to large scale.
Absence of large scale district heating networks that biofuels can 'plug' into.
The views expressed in this cuticle are those of the authors
and do not necessarily represent those of the Department of Energy,
or the Energy Technology Support Unit
1 D O Hall "Biomass Energy" 1991 Butterworth-Heineman Limited Energy Policy 19 (8) October 1991
4-96
-------
Fig 1: World Energy Use
(VVorld energy use:TOTAL 399EJ/9U67mCoe)
Oil 32%
to
u>
Coal
Biomass 14%
Nuclear 5%
Hydro 6%
Gas 17%
-------
Fig 2 : IEA Countries
Proportion of National Primary
Energy Supplied by Biofuels
Percentage
Fin Swe Atis Can Jap Den NZ SWIT It UK
-------
Hydro
(Large scale) 76%
Landfill Gas
4%
Others 1%
Wood combustion
8%
Biofuels 23%
Straw combustion
3%
Refuse combustion
8%
TOTAL renewables used = 2.1 million tonnes of oil equivalent
Figure 3. UK Renewable Energy Contribution (1990)
-------
THE UK DEPARTMENT OF ENERGY PROGRAMME
General
The Department of Energy invests in a substantial programme geared towards
harnessing 'environmentally friendly and commercially viable renewable sources of
energy'. Renewables are recognised as offering diversity of energy supply and thus
greater security of supply for the nation. Such sources are also viewed as 'sustainable'
'greenhouse gas' friendly (by recycling carbon through photosynthesis and by utilising
methane) and clean (low in sulphur). As such they offer substantial environmental
benefits over many traditional fuels.
Government has so far invested a total of £180 million (since 1979) in
renewable energy research and development projects, with approximately £13 million
allocated to biofuels. The current renewable energy programme budget is running at
£24 million per annum - a 20% increase on last financial year. Figure 4 indicates
historic spend on biofuels.
Since its inception the Biofuels Programme has sought to define the size, nature
and likely timing of the UK biofuels resource. This is a task which continues to this
day with the progressive refinement of estimates and projections. The maximum
accessible potential is believed to be very high, of the order of 200TWh/y which is
equivalent to more than half the UK's current electrical needs. The bulk of the
resource relates to crops - in particular short rotation coppice grown on land surplus to
food growing needs. Some 50TWh/y can be produced from wastes and is available at
up to 6 p/kWh.
As already discussed, only a very limited use is currently made of biofuels - of
the order of a quarter of one percent of the UK's total primary energy needs. Those
fuels which are making a contribution include landfill gas, wood (from existing
forestry), refuse and straw combustion. For these fuels the emphasis within the
Programme is switching from more fundamental studies to work geared toward
deploying the technologies. Hence, field trials and demonstrations are being worked up
where these do not exist already. The detail of the Programme is explored in the
central part of this paper. Maximum benefit is also being derived through monitoring
projects which are proceeding as a result of the Non-Fossil Fuel Obligation (NFFO), a
major market enabling measure for renewables. NFFO is an obligation placed upon
Regional Electricity Companies to purchase a defined quota of renewable energy
generating capacity. In order to secure the necessary capacity a premium price is paid
to generators with the cost being met through a levy placed on consumers. So far it has
resulted in more than 100 biofuels projects coming forward with a total installed
capacity of more than 600MWe (See figures 5 and 6).
Apart from such technical measures great effort is also being placed on
addressing other non-technical issues and barriers thought to be preventing the
widescale adoption of technically mature biofuels. Matters which are of particular
concern include uncertainty over environmental control and regulation of wastes, lack
4-102
-------
Fig 4 :UK BIOFUELS PROGRAMME
ANNUAL EXPENDITURE
2500
2000
1500
O
CJ
1000
500
0
77/78 78/79 79/80 80/81 81/82 82/83 83/84 84/85 86/86 86/87 87/88 88/89 89/90 90(91
FINANCIAL YEAR
EXPENDITURE, £k
HI
1
1
-------
9
Fig 5: Numbers of existing and planned
NFFO Projects by Area
Combustion
Landfill Gas
53
-------
Fig 6: Declared Net Capacity of existing
and planned NFFO projects
Wind 94.7 MW
Landfill Gas 83.5
Hydro 22 MW
Biogas 33.4 MW
Combustion 375.7 MW
-------
of financing, credibility of certain technologies, planning, perception and the future of
NFFO itself. In addition, for energy crops issues of land availability are central to any
substantial biofuel production programme.
Less mature fuels and technologies still need a great deal of research and
development effort as well as demonstration, if they are to yield the sought for energy
benefits. Included in this category are wood and crops where in the past emphasis has
been on growing the materials and not on use. The need within the Programme here is
to demonstrate that such crops can not only be grown commercially but that they can
also be used to produce both heat and electricity in a cost effective manner (initially
with NFFO support). Gasification and advanced thermal processing is another area
where considerable effort needs to be expended if significant advances are to be made.
The objectives of the current programme are to:
evaluate technically and economically the more important biofuel sources and
conversion processes and identify those with most promise,
demonstrate and promote those which are cost effective, paying particular
attention to environmental acceptability.
where appropriate, stimulate and support the development of better technology,
particularly where such improvement is an essential prerequisite to commercial
exploitation. .
Four main areas now make up the Biofuels programme:
1. Municipal Solid Waste: Landfill Gas; Anaerobic Digestion; Combustion and
Refuse Derived Fuel
2. Dry wastes : Dry wastes arising from industrial, commercial,
agricultural and forestry operations in the UK
excluded from the municipal solid waste category
3. Wet Wastes : Anaerobic digestion of aqueous organic wastes
including sewage, food and drink processing
effluents and animal slurries
4. Crops : Non food crops grown specifically as energy crops
eg coppice
Programme Liaison
The Biofuels Programme interacts with many other organisations world-wide in
order to both share and benefit from a much larger 'pool' of research effort. Such
liaison also allows the programme to influence the debate on regulation and other vital
non-technical issues which impact on the : .rtherance of renewable energy in the UK
and elsewhere. In the United Kingdom the energy-from-waste aspects of the
programme are closely coordinated with a complimentary programme, funded by the
DoE. A co-ordinated approach has also been developed with other Government
departments, including MAFF, particularly on the crops programme, DTI, Forestry
Commission and the Departments of Health and Transport are also involved.
4-106
-------
On the international scene, the UK . through the programme, participates fully
in the IEA (International Energy Agency) Bioenergy Agreement. There is also a very
active US/UK Bilateral Agreement to share information on biofuels and related
environmental themes. Currently, several topics are being discussed by the four
participating Government Departments (UK DEn and DoE; US Department of Energy
and EPA) for collaborative action and a number of jointly sponsored conferences are
being organised. There is also interaction with the CEC, particularly with the
Directorates on Energy (DGXVII), Environment (DGXII) and Agriculture (DGVI).
THE DETAILED PROGRAMME BY AREA
1. Municipal Solid Waste
There are a variety ways of utilising MSW and recovering energy. Within the
current Department of Energy programme there are two key routes: landfill gas and
combustion.
1.1 Landfill Gas
This area represents one of the most successful parts of the biofuels programme.
The Department initiated the earliest work on landfill gas in the UK (in 1979), and
remains in the vanguard of activities.
Approximately 50 million tonnes of municipal solid waste (encompassing
household, civic amenity, commercial and industrial waste) is produced annually in
the UK, with up to 90% currently disposed of to landfills (1991 figures). To date
landfilling has offered a cheap and acceptable means of disposal, although more
recently increased awareness of the environmental consequences of landfilling has led
to closer scrutiny of procedures and a requirement for increased vigilance. Even
allowing for greater use of incineration and municipal solid waste digestion it is
generally considered that, for the foreseeable future, landfilling will continue to play a
major role in the UKs waste management strategy. It is also thought that the potential
for utilisable quantities of landfill gas to be generated from sites will increase if landfill
practices continue to encourage fewer and larger sites, as observed since the 1980s.
This being the case there is an opportunity to exploit the positive 'energy' feature of
landfill gas generation and at the same time increase the environmental acceptability of
the process.
The Department of Energy has long perceived the benefits, both in terms of
energy and the environment, of harnessing landfill gas, and as such has had a
programme of research, development and demonstration (RD&D) investigating all
aspects of landfill gas, for the past 13 yean. The programme has focused in
particular on areas of uncertainty which may act as barriers to increased uptake of the
technology aiming to encourage the development and use of landfill gas as a fuel and to
publicise the potential of landfill gas as an economic energy resource. Six
demonstration projects were established to validate the potential in schemes ranging
4-107
-------
from electricity generation to kiln and boiler firing and played a key role in the
development of this technology. These projects established landfill gas as a
commercial technology in the UK.
It was also clear that a number of uncertainties remained and so a research
programme has been underway since 1985, managed in two halves: A Field Studies
and a Microbiology programme. Emphasis in particular is placed on the development
of techniques for improving the yield of landfill gas whilst at the same time permitting
improved environmental control. The Field Studies research programme is addressing
well design, condensate removal, and the preparation of guidelines for landfill gas
utilisation and control. The Microbiology programme aims to increase fundamental
understanding in terms of the microorganisms responsible for landfill gas production,
particularly cellulose degraders and methane producers.
Gas Utilisation Schemes in the UK.
The utilisation of landfill gas as an alternative energy resource is a proven
technology, with schemes operational in a number of countries. There are three main
options for gas exploitation :
direct burning in kilns or boilers
electricity generation
gas upgrading
There are currently 37 landfill gas utilisation schemes underway in the UK (see
Figs 7 and 8). Of these, 17 are direct use schemes such as use as landfill gas in kilns,
driers or boilers; three are pre-NFFO electricity generation schemes and the remainder
have come forward under NFFO. A further 36 schemes are planned, providing for
86.4 MW (DNC) from the NFFO supported schemes alone. It is hoped that further
schemes will continue to come forward from sites where gas collection will be required
in response to the recent legislation contained in the Environmental Protection Act
(1990), where the revenue that can be obtained from the installation of a gas
exploitation system may be seen as a means of offsetting some of the costs of installing
a gas management system.
Figures 9 and 10 indicate the energy savings for the range of uses for landfill
gas and also indicate the trend for utilisation options since 1986. The simplest option is
direct use as this involves minimal capital investment and gas pre-treatment, but it is
dependent upon the availability of a local user willing to take the gas. As this is not
possible in many cases, electricity generation from the gas is the preferred option in
the UK. This is further encouraged by the NFFO scheme previously discussed. Gas
upgrading to pipe line quality is complex and expensive and most interest in this option
has been shown by the United States where a number of purification plants have been
constructed supplying considerable quantities of methane to local gas utility networks.
4-100
-------
f
o
<0
I
m Bitumen heating
Cement making
Horticulture
Brick making
Food manufacture
(p Jy Paper making
Chemicals
Electricity
Roadstone
Metal refining
ly; :]- Effluent control
Koi breeding
(§1) Boiler
Figure 7. UK Landfill Gas Utilisation Schemes (1991)
-------
40
35
30
25
t/i
4)
e
a
¦C
o
in
•r 20
a
e
3
z
15
10
Power Gen
Kilns/ furnaces
J Boilers
Other
0li~LLL_
ll
1981 1982 1983 1984 1965 1986 1987 1 988 1989 1990
Years
Figure 8 - LAMMCOS: Number of UK Landfill Gas Schemes
1991
4-110
-------
Power Gen 122.8
Other Uses 0.9
Boilers 19.8
Kilns/Fgrnaces 42.3
Total 185.8 tcepa (thousands)
Figure 9. LAMMCOS: Total Energy Produced
200
a
c
c
o
CO
"O
C
CD
I/)
3
O
a>
a.
u
3
to
>
-------
The Microbiology Programme
The disposal of organic waste material in holes in the ground and under
circumstances where the environment rapidly becomes anaerobic, ie devoid of oxygen,
will inevitably lead to microbial activity and the production of methane and carbon
dioxide. Developing a greater understanding of the precise mechanisms involved and
how they are influenced by the immediate environment is important if the maximum
benefit is to be gained, safety standards upheld, and assessment procedures improved.
Microorganisms are responsible for the degradation of biological material and so a
programme aimed at understanding which organisms are responsible, what Conditions
they prefer, and how the landfill does or could satisfy these conditions has been
running for 4 years, with a budget of just under £1 million. Seventeen projects have
been initiated, with eleven already complete and further details and reports can be
obtained from the Renewable Energy Enquiries Bureau at ETSU.
The Field Studies Programme
The aims of the field studies programme are:
to define with greater certainty the overall technical and economic potential of
the UK landfill gas resource
to establish an extended database for monitoring and use with the UK landfill
gas resource model
to investigate methods of landfill gas enhancement and optimisation
to investigate more sophisticated and effective gas abstraction and treatment
technology
There are currently 5 projects in this area and further details are available from
the Renewable Energy Enquiries Bureau. Two of the nine energy efficiency projects
previously mentioned will now be discussed in some detail to highlight how the projects
have operated and what they achieved.
The feasibility of generating electricity for supply to the UK national
distribution network , from landfill gas fuelled spark ignition engines was investigated
at Stewartby landfill site, operated by Shanks and McEwan (Southern) Ltd, in 1987.
The project aimed to show that remote or rural landfill sites without easy access to
direct consumers of landfill gas could also capitalise upon the energy potential of
landfill gas. The project cost £418,500 and achieved savings of £158,000 per annum
offering a payback period of 2.4 years (1987 prices). The cost of the compressor was
excluded from the investment cost because this would be needed anyway if the gas was
simply flared. If gas control was not previously a pre-requisite and if existing gas
collection and well systems were not in place a payback period of the order of 4.7 years
(average exported tariff 2.75p/unit) was estimated.
Three 275 kW spark ignition engines fuelled by landfill gas coupled to air
cooled generators were installed in a single -roomed engine house containing control
4-112
-------
panels and the lubricating oil tank. Due to the lower calorific value of landfill gas, the
rating quoted was 11% below the engine's normal rating when operating on natural
gas. The units were designed to operate unattended, requiring comprehensive
interlocks to trip the units in the event of low gas pressure, spit back in the carburettor,
low oil pressure, high engine temperature or engine overspeed. All engine trip circuits
were linked by telephone to a 24 hour call out system. Both vertical and horizontal
wells were used to extract the gas and knock out drums for removal of water were
located in polyethylene lines from the landfill site. Gas was compressed after filtration
using a single constant displacement vane-type unit manufactured by Hammond
Engineering and driven by a 45kW electric motor (rated to supply 680 m^ /h) against
1.3 bar g. The compressed gas then passed through an aftercooler, baffle water
separator, chiller and fine filters before being supplied to the adjacent engine house.
Any surplus gas was flared.
Electricity was supplied at 415V for in-house use with the main export power
stepped up to 6.6kV via a single transformer for use by the local London Brick
Company. Any residual power was then exported to the local grid at 33kV.
Plant Performance. The plant operated almost continuously from February
1987. There were problems associated with a failure in the oil supply, resolved by a
replacement pump, and also three compressor failures resulting from water getting
through the water separation systems, requiring compressor rebuild or replacement in
each case. Apart from two major service shut-downs the generators ran almost
continuously, and most of the minor faults were electrical and resolved by adjustments
to the equipment. Occasionally the units were tripped by low methane content. After
22,000 hours of service the engines were in generally good condition with only minor
problems of wear.
Case History 2.
The second case history is the generation of electricity from landfill gas using
gas turbines. A 3.65MW Centiax gas turbine (model CX350 KB5 powered by a
General Motors Allison 501 KB5 single shaft gas turbine which runs at 14,250 rpm)
was installed in 1987 to generate electricity for direct export to the local grid. At the
plant the gas was scrubbed and then passed through a centrifugal blower before being
compressed in two Belliss and Morcom WH56N compressors. Each compressor was
rated at 55% duty, although each was capable of delivering 70% of the total gas
requirement when operated with the blower. Before delivery to the turbine the gas was
cooled and superheated to control condensation of hydrocarbons. The drive to the
Brush 6.125 MVA generator was taken through a step down gearbox to 1,500 rpm.
The generator output was exported to the local grid at 1 IkV via a i.8 km underground
cable.
Plant Performance. Start up of the system indicated many teething problems.
During the plant acceptance trials the automatic condensate return valves were not
functioning. Removal and cleaning.rectified this problem. During commissioning one
of the compressors failed as did the replacement unit. Consequently initial operation
4-113
-------
was undertaken using only one compressor supplemented by the gas blower. Under
these conditions it was possible to raise the turbine output to about 2.7MW. During
early running of the compressor the cylinder head and valves suffered fouling by
chloride salts and hydrocarbons, heavy corrosion also became apparent in the stainless
steel flexibles connecting to the compressor. The probable cause was traced to the
formation of hydrogen chloride through reaction of sodium hypochlorite with
hydrocarbons. Sodium hypochlorite and sodium hydroxide were being added to the
scrubber liquor to remove any traces of hydrogen sulphide present in the landfill gas.
Dosing with sodium hypochlorite was subsequently stopped following advice from the
manufacturers.
Problems of 'belling' had been reported on similar plant in the USA and so
replacement ends of the turbine fuel manifold were manufactured. Due to fracture of
the gas injection nozzles, all nozzles were replaced and subsequently redesigned.
Problems were also encountered with the scrubber control panel and this was replaced.
As a consequence of these problems the system operated with an availability of 85%
and at a reduced average output of 58%. However during June 1989 - May 1990 the
system ran with an availability of 95% while operating at 79% of rated output. All of
the monitored exhaust emissions were lower than the limits allowed for municipal waste
incinerators in the UK, except on one instance when the HC1 level emitted marginally
exceeded the limit of 250 mg/m^. The noise level at a distance of 50m was inaudible
above the total background noise level, even at night.
The investment cost was £1.946 million (1986 prices) with savings of £176,780
giving a pay back period of 11 years. It should be recognised though that this is based
on monitoring during the initial teething stage. It was estimated that if the plant had
achieved target generation with the original electricity tariff the payback period would
be reduced to 4.5 years.
Summary of Results From The Landfill Gas R&D programme
As previously mentioned several research projects are now complete. The following
outlines a selection of the achievements:
1. Development of a computer model to estimate the UK national landfill gas
resource.
2. Evaluation of well design concluding that horizontal wells were most effective
in areas of high leachate levels.
3. Evidence to show that the application of a discontinuous polyethylene cap to
previously uncapped areas increases the landfill gas content of the gas
abstracted.
4. Demonstration of electricity generation from landfill gas using reciprocating
engines and gas turbines.
5. Evidence to show that some of the bacteria involved in the landfill degradation
process are specific to the landfill environment, endorsing the notion that we
cannot just apply existing knowledge from similar systems eg anaerobic sewage
sludge digestion.
4-114
-------
6. Techniques for, and experience of, setting up iaboratory scale model landfill
systems.
7. Evidence to suggest that fungi do not play a major role in refuse degradation.
8. Evidence to show that some species of methane producing bacteria can be
protected from external environmental fluctuations by co-existing within
protozoa.
9. A comprehensive review of microbiological methods applicable to landfill
studies and identification of areas of weakness.
10. A review document discussing the possible effects of recycling on waste
composition and gas production.
11. A review of the potential for applying DNA technology to the study of
landfills.
A number of projects are currently underway or at the final planning stages and
it is hoped that information on the following will be available shortly:
1. Review of demonstrated prime mover engine technologies, mainly focusing on
the UK but also looking overseas. The study will include corrosion, erosion and
wear; the impact of legislation and emissions.
2. A feasibility assessment of the potential for developing a bioreactor cell rotation
landfill.
3. An assessment of various enhancement techniques on the production of landfill
gas at the Brogborough Landfill Test Cells.
4. An investigation into cellulose degradation in landfills.
1.2 Combustion of Municipal Solid Waste fMSW)
Estimates of annual UK waste arisings (domestic, industrial, commercial and
straw) suggest that the combustible fraction has an energy content equivalent to some
28 million tonnes of coal (Mtce) with an energy value in excess of £1000 million.
Some 3.5 Mtce is estimated to be economically accessible at present energy prices,
although less than 10% of this is currently realised.
In the UK Municipal waste combustion is dominated by mass burn incineration,
with 34 plants providing a disposal route for 8-10% of UK arisings. These all date
from the 1960s and 70s and were built by chiefly as a means of volume reduction prior
to transport to landfiiling. Energy was not an issue, and only 5 of the plants are
equipped with energy recovery. Most of the plants are approaching the end of their
serviceable life and, with the need to comply with recent comprehensive European air
pollution regulations by 1996, it is anticipated that the majority will close rather than
be retrofitted with the necessary abatement equipment.
Incinerators capable of meeting modem environmental requirements are
considered to be proven technology. However, in the UK context of low cost landfill
4-115
-------
they represent a high cost disposal option. DEn studies have therefore focused on the
impact and maximisation of energy recovery to enhance project economics. CHP and
district heating have been identified as preferred options. However, the UK markets
for these are limited and hence, with the introduction of the Non-Fossil Fuel
Obligation, the focus has moved to electricity production. Seven retrofit and new
schemes have been proposed since the initial obligation was made in 1990, including a
1.5 Mte/yr 102 MWe plant for London (which will be the largest such scheme in
Europe). In the absence of other fiscal measures, the NFFO is widely regarded as the
major factor which will determine the future scale of mass bum incineration in the
medium term.
In parallel with the economic studies, field trials have been undertaken at two of
the existing heat recovery incinerators to evaluate the potential of mechanical screening
to improve combustion and reduce the burden on the downstream gas cleaning
equipment. Screening (at 50mm) was found to increase the calorific value (CV) by
30%, reduce the bottom ash by 30% and halve the carbon in ash losses. Energy
recovery per tonne of sorted waste increased by over 40% compared with untreated
refuse, reflecting both the increased CV and an overall improvement in boiler
efficiency. Preliminary economic assessment indicates that pre-screening could yield a
20% saving on net disposal costs. Long term operational assessment is now required.
The principal aim of the programme, however, has been to assess alternative
options to mass burn incineration for the typical scales of urban waste arisings
(150-250,000 te/yr). This has concentrated on waste processing and the utilisation of
refuse derived fuels (RDF), particularly the utilisation of high processed densified
RDF as a substitute for lump coal on industrial solid fuel fired boilers.
Refuse Derived Fuel
The UK development of dRDF was initiated by the Department of Environment
in the mid-1970s against the backdrop of a perceived energy crisis and a shortage of
landfill capacity. Following initial development work a number of second generation
dRDF plants were established in collaboration with Local Authorities, ostensibly as
recycling centres with the RDF representing the reject paper fraction. In the event,
markets for the recovered materials failed to develop and the dRDF became the
principal product. DEn involvement in the utilisation of dRDF as an industrial fuel
substitute commenced in 1981 with the sponsorship of a series of industrial user trials.
Some 15 individual projects have since been initiated at a cost of £1.5 million. It
rapidly became obvious that, due to the higher volatiles content, lower bulk calorific
value and greater fouling potential of dRDF compared with coal, significant boiler
modifications would be required in order to achieve fuel substitution. Much of the
further work was aimed at understanding the fundamental combustion characteristics of
dRDF, its handling and storage characteristics, the most suitable types of furnace for
dRDF firing and the nature of the boiler modifications required. The chain grate stoker
shell boiler was identified as particularly amenable to dRDF firing provided provision
was made for over-fire air to ensure adequate bum-out of the volatile components and
also for temperature control and soot blowing at the rear of the fumace to minimise
boiler fouling. The programme was closed in mid-1991 when sufficient data for an
4-116
-------
accurate assessment of the economics of dRDF production and utilisation had been
amassed. The work showed that dRDF can now be produced and used without
technical problems. However legislation requiring more comprehensive abatement of
emissions from dRDF than from coal was introduced subsequent to the 1990
Environmental Protection Act. This renders the use of dRDF uneconomic against coal
and other industrial fuels in most circumstances. However, four schemes have been
accepted under the Non-Fossil Fuel Obligation and will commence electricity
generation in 1991.
European policy is now that waste minimisation and materials recycling are
preferable to incineration, with landfilling being a waste disposal option of last resort,
and the UK has since introduced a recycling target equivalent to about 25% of
municipal waste. Means of progressing towards this target are being investigated,
including source separation and "blue box" systems. However, centralised processing
for materials recovery, coupled with on-site energy production, may provide an
economic option for waste disposal. Such integration, termed here Resource Recovery,
is now being considered under a new programme strategy building on the success of the
dRDF programme. Initial economic studies are currently underway in collaboration
with major boiler manufacturers to assess the viability of this option, ahead of
implementing the R,D&D elements of the programme, and will draw on the cRDF
(coarse RDF) and MRF experience in the USA.
Industrial and Commercial Wastes
It is estimated that industrial and commercial present the largest single waste
energy resource, equivalent to 15.3 million tonnes of coal each year. This includes
specialised wastes, produced as a result of a particular activity or process, for example
scrap tyres and hospital waste. Most, however, is general industrial waste (GIW),
made up largely of paper, cardboard, wood and plastics. This differs in composition
from raw domestic waste and is an inherently better fuel, being generally less
contaminated and with lower moisture and ash contents (both generally less than 10%).
In 1981 a review revealed that the vast majority of this waste is traditionally
disposed of to landfill. A series of projects were therefore established under the
Department of Energy's (DEn's) Energy Efficiency Demonstration (ED) Scheme to test
out opportunities for their use as an industrial fuel. Reviews of the results revealed that
the projects, especially those associated with GIW, had experienced many technical,
organisational and contractual problems. Nevertheless a number of opportunities were
identified for producing low-cost systems with improved characteristics which, if
developed effectively, could prove attractive to industry. Subsequently, a strategy for
industrial and commercial wastes as fuel was adopted by the Department of Energy
under its Biofuels R&D Programme. The following summarises the main areas
investigated:
Waste Preparation. Handling and Storage Technology. Demonstration projects have
shown .that shortcomings relating to the fuel processing, handling and storage have
proved a major barrier to the uptake of this technology. The degree of processing
4-117
-------
necessary for the use of industrial and commercial waste as fuel is largely determined
by the requirements of the combustion system used. Major projects on waste handling,
storage and preparation, together with shredder development studies and long term
storage trials have therefore been undertaken.
Combustion Technology. For the 'on-site' market with outputs in the range of
1.5-3MW thermal, waste burning shell boilers, biomass pre-combustors or retrofit
waste firing on existing boilers offer the best prospects.
In 'off-site' applications there is currently considerable interest in both CHP and
power generation schemes. For this market, waste burning boilers, composite boilers
or water tube boilers offer the best prospects.
Waste Burning Boilers. A 1 te/h waste-fired CHP facility is currently being evaluated
under the DEn programme and large output installations can be built from a number of
these units. The major limitation with currently available vertical shell units is that the
maximum working pressure of the boiler is limited to lObar. New designs operating at
higher pressures are therefore being considered.
Composite boilers generally require only coarse waste preparation and operate
at higher pressures than shell boilers, with the option to provide super heat giving
improved power to heat ratios for CHP and Power Generation. Despite these inherent
advantages, no systems are known to have been installed for waste firing.
Water Tube Boilers. Numerous combustion systems are available which, in principle,
allow these boilers to deal with a variety of wastes (mass bum grates, fluidised beds,
chain grates etc). A provisional design for a 5.5MWe power generation facility is
currently being considered under the DEn programme. This is based on reciprocating
grate technology with a bulk breaker, magnetic separation and a continuous belt storage
bunker fuel handling system.
A particular problem exists with electricity production in small scale systems
due to the relatively low performance of small scale steam turbine generators. A desk
study of the potential of reciprocating steam engines and Sterling engines for electricity
production is being undertaken as a complementary pan of the programme, with a view
to potential collaboration in longer term development.
2. Dry Wastes
Specialised Industrial Wastes (SIW>. SIW includes hospital waste, waste wood,
scrap tyres and many other opportunistic waste which may be utilised as an energy
resource. Initially, the 'on-site' market was considered the most important for the
utilisation of this resource, but more recently the potential for 'off-site' use has grown
as major waste disposal companies have become interested in its exploitation. Notably,
a 40 MWe private venture tyre-fired power station in now under development under
NFFO, based on the US experience of multiple modular incinerator technology.
4-118
-------
Hospital Waste. A major driver in the development of SIW utilisation has
resulted from changing legislation affecting hospital waste incineration. Traditionally,
this has been carried out 'on-site' at some 900 hospital locations in England and
Scotland, mainly using rudimentary batch fired technology. Legislation introduced
subsequent to the 1990 Environmental Protection Act now require such plant to meet
strict performance standards. Further legislation proposed within the European
Community may increase control yet further and few, if any, of the traditional
incinerators could be retrofitted to comply with these requirements. New capacity will
therefore be required, creating a potentially buoyant market. Health Authorities in
particular are considering new group incineration schemes, financed internally, joint
funded, or owned/operated by private waste contractors. Energy recovery is feasible
but, because of the high disposal fees available is not a necessary prerequisite for
economic viability. However, the market should prove fertile for the development of
monitoring systems and advanced small scale pollution abatement technologies.
Straw. Annual straw production in the UK is estimated to be 17 million tonnes,
of which up to 7 million tonnes are either burned in the field or ploughed in. The
surplus straw, equivalent to 3.5 million tonnes of coal, is potentially available as a fuel
and already some 02 M tonnes are consumed by farmers for farm heating, mainly
based on simple batch fed whole bale burners. The potential for industrial and
commercial use, however, is limited by the delivered cost of straw, which generally
exceeds that of coal on a specific energy basis. Research towards reducing this cost
have focused on densification of the straw to reduce transport costs. Other fundamental
studies on the cohesion within a wafer and on the effect of variations in swath density
are underway but it is now considered unlikely tha: a sufficiently robust system can be
developed for in-field use. However, the basic system shows considerable promise as a
static densifier for a wide range of organic and waste feedstocks.
Case Studv: Poultry litter. Fibropower Ltd. are currently building a poultry
litter fired power station at Eye in Suffolk. Construction of the power station began in
Spring 1991 and it is to be commissioned in the Summer 1992. The project involves
the collection of about 130kt/y of poultry litter within a 40km radius of the station
which will be burnt to provide 12.6MWe of electrical power for export. The furnace
ash and fly ash from the precipitator are high in phosphate and potash and will be
removed from the site in bulk, for use as a base in the manufacture of agricultural
fertilizers.
Originally, Fibropower planned to operate a CHP scheme generating about
5MWe but owing to commercial difficulties in selling the steam the Company decided
to opt for an electricity only project. The project subsequently applied for, and was
successful in obtaining, a contract under the first tranche of the NFFO. The project
also received £750k from the European Commission, which included £12k to cover the
cost of some of the specialised monitoring equipment. The DEn, provided £78k to
cover the cost of monitoring which is being undertaken by FEC Consultants Ltd.
4-119
-------
3. Wet Wastes
This part of the programme focuses on the potential for the production and
utilisation of energy from biogas produced as result of the anaerobic digestion of wet
waste, including sewage sludges; food and drink processing effluents: and animal
slurries. Approximately 70,000m3 of wet organic sludge is produced as an unwanted
by-product of sewage treatment each day in the UK. Of this, 30% is disposed of
untreated to farmland or landfill. The remaining 70% is disinfected, deodorised and
stabilised by anaerobic digestion in large digesters before spreading on selective
farmland or disposal at sea ( the latter permitted only until 1998). Such systems are
commercially available and a proven technology with the sewage derived biogas
generally used on site.
Food and drink processing effluents suitable for anaerobic digestion include
both high solids food residues and dilute factory washwaters. A number of system
designs are available, although digestion of such wastes is considered to be a
developing technology in the UK.
Anaerobic digestion of farm slurries i.e. cattle, pig and chicken is still an
emerging technology in the UK, with only one operational scheme recovering energy.
The low uptake is a function of the economics of the process and the challenge in this
area is to develop a low cost 'rugged' digester which farmers can afford to use.
4. Crops
Wood as a Fuel
The key part of the crops programme is "Wood as a Fuel". This work began in
1979 and considered a promising but uncertain technology for the future. From then
until the present Department of Energy funded research, with a value of £1.7 million,
has demonstrated the potential for wood fuel to become a major renewable energy
resource in the UK, and the means by which this might be achieved. Fuel wood
supply from conventional forestry operations have been considered from harvesting
through to comminution, transport and storage pre- utilisation. The various
technologies have been researched and the associated incurred costs determined, using
a work study approach, enabling fuel wood strategies to be defined using a flexible
computer programme. Fractionation of residue derived fuel chips to allow removal of
higher value 'white* chips, to subsidise the production of lower grade fuel chips, is also
being researched with the aim of lowering the costs of fuel production.
Research into arable energy forestry was limited to two projects during this time
which demonstrated the potential for coppice derived fuel wood. During this period a
single demonstration scheme was set up at the Tormore distillery where although the
steam boiler based on wood fuel was a technical success, the company ceased
operations.
4-120
-------
The Wood as a Fuel programme has grown recently as a result of studies
suggesting that fuel wood from conventional forestry operations might be possible at
prices competitive with those of fossil fuels. This has coincided with the food over-
production in the UK requiring farmers to diversify to non-food crops; planting grants
being made available, during 1991 and increasing concerns about local and global
environmental issues raising the profile of producing and using wood as a fuel.
Considerable effort is now being expended towards market development in particular
enabling demonstrations of the relevant technologies for wood fuel uptake to be carried
forward. It is considered that if a suitable market could be developed, the
conventional forestry wood fuel source would initially be used to satisfy demand, with
arable energy forestry coming- on stream later, as the market expanded. Arable
forestry though is seen as holding the greatest potential for wood fuel.
The aims of the Wood as a Fuel programme are to:
take the research to the demonstration phase
aid market development through demonstration
assess the economics of wood fuel supply and use
carry out underpining research
investigate the environmental implications of fuel wood provision and use.
For the purpose of this paper, wood fuel will be considered to be available
from two main sources, existing "conventional" forestry operations, or short rotation
coppice grown specifically as a fuel crop. Wood fuel derived from conventional
forestry is considered as an additional available crop and not a residue or waste. Waste
wood encompasses offcuts from manufacturing industry or timber from building
demolition. They are considered separately for a number of reasons:
the focus of the Wood as a Fuel programme is crop utilisation rather than waste
recovery, and this has significant influence on the economics and supply
infrastructure;
waste wood is often chemically treated or otherwise contaminated leading to
potential problems associated with emissions during utilisation;
capital equipment is required to produce the fuel in a cost effective manner
the resulting wood fuel can subsidise the production of the stem wood.
Fuel Wood from Conventional Forestry.
Forest management techniques are aimed at producing straight knot-free stemwood,
which is achieved by planting trees closely together with thinning as required. In
unmanaged woodland no thinning occurs and natural re-growth of young trees results
leading to "derelict" woodlands. Trees tend to no longer grow straight due to,
amongst other factors, light competition, and they also tend to grow at less than their
maximum incremental rate. Consequently the standing value of the timber falls and
utilisation options are limited. Wood for use as fuel can be derived from the operations
required to bring such derelict woodland back into production, from essential forestry
thinning operations and also from final harvests.
4-121
-------
JL Final harvest The raw material available for production of fuel
following British 'shortwood' final harvesting is the branches and small diameter tree
tops (the 'brash'), which often account for more than 30% of the tree biomass, and is
either left on site to rot or is removed prior to subsequent planting. Re-entering the
site post harvest is obviously possible but involves doubling the wood handling. This
obviously is not efficient and results in a high delivered price of fuel - £2.18 to
£2.33/GJ. It does however mean that the only extra item of capital equipment required
is the chipper.
Studies funded by the UK Department of Energy have demonstrated the
increased efficiency of 'one pass' integrated harvesting operations, where the whole
tree is removed from the forest after mechanical harvesting and subsequently separated
to stemwood and fuel wood at the roadside. As the 'one pass' implies the material is
only handled once and greater processor and chipper productivity is facilitated in the
more stable environment of the roadside. The increased productivity and efficiency of
operations reduces the delivered price of fuel wood to, in the best circumstances,
£1.17/GJ. Increased productivity of stemwood also results, and since the cost of
harvesting, transport and product separation are shared proportionately between the
harvesting and the wood fuel operations, an effective subsidy of the costs of producing
stemwood is achieved. The major drawback to such integrated harvesting though is the
high capital equipment costs.
Zj. Essential Thinning. The second potential source of fuel wood is from
essential thinning operations, where whole tree chipping is the cost effective option.
This approach has been demonstrated at Thetford forest in the UK and indicated high
productivity with a delivered price of wood fuel in the region of £1.79/GJ. The
capital equipment costs are once again high for this procedure.
X. Derelict Woodland. Returning derelict woodland back to production
may involve any combination of the above techniques, and it is likely that a fuel wood
market would be the only one available to timber from badly neglected woodlands.
Whilst utilisation of the brash undoubtedly increases the efficiency of harvesting
and thinning, the mechanical requirements lead to high capital costs and consequently
will act as an impediment to adoption of these practices until a fuel wood market exists.
The development of such a market will be dependent upon the production of fuel at a
price competitive with fossil fuels which still remains a challenge. However when it
does is should be long term and stable, permitting a return on capital expenditure, so
offering a relatively attractive proposition.
One problem with using conventional forestry, not previously mentioned, is
the fact that the amount of wood fuel available is constrained by the forestry resource,
which is only 7% of UK land area. It is in tackling this issue that the role of arable
energy forestry can have most impact.
4-122
-------
Arable Energy Forestry
In order to increase the amount of wood available, adoption of traditional
coppicing methods for modem production is possible. Research has demonstrated that
unrooted willow or poplar cuttings can be planted in closely spaced rows at
approximately 10,000 trees to the hectare. The ground should be prepared and cleared
of weeds in the autumn, planted in the spring and then harvested at two to five year
intervals after initiation of coppicing. A crop yield of 10-15 dry tonnes of fuel per
hectare per year has been demonstrated. Such • plantations are best established on
agricultural land and offers a good option for utilising land currently in food over-
production. In the UK, coppice, as a designated non-food agricultural crop is eligible
for 'set aside' payments (a scheme to encourage farmers to take land out of food
production), and as added inducement is now included in the Forestry Commission's
Woodland Grant Scheme which provides planting grants, as mentioned earlier. By
using agricultural land in this way, as well as marginal areas, coppicing can provide
an innovative and financially beneficial form of diversification and so arable coppice
should not be seen as an alternative to forestry. A further attraction of arable coppice
is the environmental acceptability of the process:
The carbon dioxide given off during the burning of wood has previously been
absorbed during the trees' lifetime, and so is a recycling process - there is not
net gain in carbon dioxide levels.
When compared with coal burning there is also the added benefit of reducing
acid rain.
Poplar and willow were demonstrated to be the best two species for arable
energy forestry in the UK, in agreement with research from other countries. Other
species investigated gave poor establishment and high rates of failure on coppicing or
poor yields. Until the canopy is large enough to cut out light and suppress growth on
its own, or the root system is large enough to outcompete weeds for water,
maintainence of a weed free plantation is essential. Prevention of rust disease which
can cause premature defoliation leading to reduced crop yields is also important. At
the end of the first growing season, the shoots produced are cut back to initiate
coppicing, and because of the clonal nature of this crop, the shoots produced can
themselves be used to produce cuttings for subsequent planting. By planting fields in
rotation annual harvests and hence income can result. Harvesting is carried out in the
winter any time after leaf fall (October/November) to leaf set (March/April) and a
prototype tractor drawn harvester has been developed as part of the Department of
Energy's research programme. This system uses contra-rotating augers to gather the
stems which are cut close to the base by a circular saw. With further development it is
anticipated that one hectare of three year old coppice per day could be harvested.
Stems are collected in bales of 300kg and stored until required. Baled storage is
important because this prevents microbial breakdown, in turn avoiding the need for
expensive storage containers. A further advantage during the summer months is air
drying of the bundles which leaves the wood more amenable as a fuel when it is
ultimately removed for chipping and use. Another aspect of the Department's
4-123
-------
programme is monitoring establishment, growth, harvesting, storage, combustion, and
effect on the landscape in terms of environmental effects.
It has been estimated that the cut stool will remain in production for up to 30
years before replanting becomes necessary, although yields will decline towards the
end of the life period. Although costings will vary on a case by case basis the basic
costs associated with arable energy forestry encompass;
fencing
ground preparation
cuttings
harvesting
chipping.
A project has been established (1992) in which a series of farmers are working
together as a cooperative, producing wood fuel from arable energy forestry. Crop
performance is being monitored, as are:
the integration of the crop with other on-farm activities
environmental effects and cash flow
profitability
It is hoped that the project will help bring the product into the market, by
assisting with crop production and its subsequent availability and sale.
Utilisation of Fuel Wood
Wood makes a good combustion fuel, however when burnt in open fires or
wood burning stoves smoke and volatile pollution can be produced due to the batch
feeding process involved. The preferred option therefore is for wood chips to be burnt
in modern, refractory lined combustors with provision for secondary air which give a
highly efficient and clean bum, if high combustion temperatures are maintained. Such
wood chip fired plant can generate warm air, hot water for wet heating systems, or
raise steam for process or electricity generating purposes. In the longer term thermal
processing techniques to produce gaseous or liquid fuels from wood may well develop
and become a UK option.
Wood combustion does have its problems though, the two main ones being size
(wood has a low bulk density) and cost (typically twice that of equivalent coal fired
plant, and three times that of oil or gas.) Consequently wood combustion is likely to
be more suited to industrial or institutional users.
4-124
-------
$r-
FUTURE EMPHASIS
UK energy production is increasingly moving towards decentralised, low cost
natural gas fuelled plant, particularly for power generation, and only in specific
circumstances can biomass derived energy compete. Economics have traditionally
represented a major barrier to the commercial development of renewable energy and so
the creation of an internal market for renewables via NFFO, as an incentive to
commercialise renewable energy technologies, has been an important aspect of the
Government's energy policy. In the absence of large scale commercial heat or district
heating networks, this emphasis has been on electricity generation. By using the
proceeds of NFFO to pay higher tariffs for the electricity generated the financial
viability of schemes from renewable and other non fossil fuel sources has improved.
Since its inception a total of approximately 600MWe installed capacity has been
contracted to supply (this figure includes electricity generated from hydro and wind
farms).
The UK, in line with wider European policy, is focusing on waste minimisation,
materials recycling and energy recovery, and tightened controls over waste disposal as
a means of implementing its Integrated Pollution Control commitment. Consequently it
is envisaged that industrial waste at source will be reduced and that recycling in the
domestic and commercial sector will increase. In order to harness market forces to
more effectively encourage waste minimisation, the Environmental Protection Act has
established a strict environmental regime for the disposal of waste both to landfill and
incineration, the two currently viable options for disposal. This will be manifested in
increasing disposal costs which will provide a strong incentive for waste volume
reduction. It is thought that landfllling costs, in particular, will rise relative to other
forms of waste treatment and disposal so that incineration and 'resource recovery'
centres will be less dis-advantaged. Although transfer stations enabling waste to be
transported over long distances have increased in recent years it is still anticipated that
incineration will increase in popularity, particularly in cities where available void is
limited.
Refuse derived fuels were originally envisaged as offering an intermediary
between landfill and incineration in terms of costs. Research concentrated on the
production of pelletised or densified RDF (dRDF) which was targeted as a potential
coal substitute for use on small scale boiler systems, either to generate process steam
and/or provide space heating. It has failed to enter the waste disposal market at a
significant level largely because of the failure to secure viable markets for the fuel,
and its prospects are weak since recent emissions requirements have meant that any
possible economic advantage of dRDF over conventional fuels has been eroded.
In order to promote greater recycling and reuse of materials the Governments
aims are to encourage both the supply of, and demand for, recycled materials. In the
domestic sector, where householders do not pay directly for the disposal of their
waste, it is hoped that a system of 'recycling' credits will be introduced whereby the
savings in landfill costs will be passed on to waste collection authorities and voluntary
recycle organisations enabling recycling to become more profitable. Recycling in the
4-125
-------
industrial sector is already more successful and the rising costs of waste disposal are
likely to act as incentive to improve further.
In terms of energy recovery the increasing environmental pressures are
generally acting as a spur. Although the anticipated increased costs of landfilling,
legislative requirements and the decreasing availability of void is likely to lead to a
reduction in landfilling and an increased role for incineration (particularly for large
conurbations), for the foreseeable future landfill in the UK is likely to play a major,
although less significant role than at present in waste management practice. The
challenge therefore is to increase the control of gas production at sites thereby
maximising both environmental control and facilitating effective energy recovery.
Incorporation of an energy recovery component in the incineration and resource
recovery processes will also be targeted. Anaerobic digestion of farm slurries and
manures with energy production is still an emerging technology, in the UK, and so
progress of the single scheme accepted under the first NFFO tranche will be eagerly
followed.
A further possible boost to energy recovery ventures for other waste materials
may result from recent legislation in the UK which has banned the disposal of sewage
at sea after 1998, and also straw burning in fields effective this year, meaning that
alternative disposal mechanisms will need to be found. However preliminary
assessments so far suggest that this will make only a modest contribution.
No commercial projects currently exist in the United Kingdom for making use
of energy crops although wood from conventional forestry is commonly used.
However, the massive potential which could conceivably be derived from this resource
is likely to encourage a substantial growth of interest. Longer term work on a range of
energy crops and complementary technologies may be possible producing solid, liquid
and gaseous fuels. In the short term, the Department of Energy's research
programme is pioneering effort in terms of the growth, harvest and use of coppice
wood fuel through a fanning cooperative with the hope of providing market stimulus
and incentive to producers and manufacturers. The effect of NFFO in this area has
also been limited to date due to the technology being at the early development stage.
In summary the situation for commercialisation of electricity generating
renewable energy schemes looks promising in the UK, and a contribution of 600MWe
- 1000MWe by the end of the century looks attainable.
ACKNOWLEDGEMENT
Thanks to Arabella Hardcastle for the preparation of figures used in this paper.
4-126
-------
SESSION V: BIOMASS EMISSION SOURCES AND SINKS
Robert Dixon, Chairperson
THE CARBON BALANCE OF FOREST SYSTEMS: ASSESSING £££
SI!£filii£S_aJlilZLJ^USUUSSil£itQ!_OTS&&ililX££&2—
ANP FLUX
by: Robert K. Dixon
U.S.EPA Environmental Research Laboratory
Corvallis, OR 97333
Jack K. Winjum
National Council for Air and
Stream Improvement
U.S.EPA Environmental Research Laboratory
Corvallis, OR 97333
Paul E. Schroeder
ManTech Environmental Technology, Inc.
U.S.EPA Environmental Research Laboratory
Corvallis, OR 97333
ABSTRACT
5A
Forests play a major role in the Earth's carbon cycle through
assimilation, storage, and emission of CO2. Establishment and
management of boreal, temperate, and tropical forest and
agroforest systems could potentially enhance sequestration of
carbon in the terrestrial biosphere. A biologic and economic
analysis of forest establishment and management options from 94
nations revealed that forestation, agroforestry, and silviculture
could be employed to conserve and sequester one gigaton (Gt) of
carbon annually over a 50 year period. The marginal cost of
implementing these options to sequester 55 Gt of carbon would be
approximately $10/ton.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication.
5-1
-------
INTRODUCTION
The accumulation of greenhouse gases in the atmosphere due to
deforestation, fossil fuel combustion, and other human activities
may have begun to change the global climate (12,26,38). Given our
current understanding of global carbon sources and sinks, the
prospect for managing the terrestrial biosphere to alter the
carbon cycle and reduce the accumulation of greenhouse gasses
appears promising (12,21,56).
Forest and agroforest systems play a prominent role in the
global carbon cycle (26,47). Forests contain over 60% of the
terrestrial above-ground carbon and approximately 45% of the
terrestrial soil carbon (1,53). In addition, worldwide forests
account for approximately 90% (90 gigatons, Gt) of the annual
carbon flux between the atmosphere and terrestrial ecosystems.
Based on preliminary estimates, application of forest management
and agroforestry practices to stimulate biomass productivity on a
global scale could potentially sequester or conserve several
gigatons of carbon annually (12,21,56).
Agricultural systems also play a significant role in the
global carbon cycle (7). They contain about 12% of the world's
terrestrial soil carbon, and conservation of this pool is
essential to sustained crop productivity and decreasing CO2
emissions (7,26,27). Many agricultural practices have been shown
to increase soil carbon content by increasing carbon sequestration
and/or reducing the loss of carbon. Practices such as reduced
tillage, crop residue incorporation, field application of manure
and sludge, and rotations using cover crops or leguminous crops
store more carbon than conventional technology (27).
Recognizing the prominent role of forest and agroforest
biomes in global ecology and the global carbon cycle, participants
at the United Nations Conference on Environment and Development
(UNCED) developed a set of forest principles (20). The forest
principles have several purposes including:
* slow deforestation
* protect biodiversity
* stimulate sustained forest
management and productivity
* address threats to the world's forests
Of primary concern in shaping these objectives were several
proposals in the past year for an international convention,
charter, protocol, or other agreement to maintain, manage, or
protect boreal, temperate, and tropical forests (31).
In 1989, delegates at the Noordwijk Ministerial Conference
recognized the role of forests in transnational environmental
issues, including global climate change, and stimulated interest
in accelerated forestation and sustainable ecosystem management
-------
options (35). The conference recognized the significance of the
observed increases in atmospheric carbon dioxide and established a
provisional net world-forestation goal 12 million hectares (ha)
per year, which is to be reached by the year 2000.
The potential role of forest establishment to increase carbon
sinks and stimulate sequestering of atmospheric carbon has been
considered by a number of authors (2, 16,17, 22, 44) . These analyses
have emphasized the major forest regions on a continental basis,
especially within tropical latitudes. Though preliminary, these
analyses have shown that forest and agroforest establishment and
management appear to have significant promise for contributing to
global carbon sequestration and conservation (38,54). At the same
time, implementation of these practices has the potential to
provide a continuous flow of forest-based goods and services
(15,19).
Given the scope of the science-policy needs regarding global
forests, the global carbon cycle, ana climate change, two specific
objectives were established in this assessment:
1. Identify promising technologies and practices that
could be utilized at technically suitable sites to
manage forest and agroforest systems to stimulate
biomass productivity and sequester atmospheric
carbon.
2. Assess economic potential, specifically costs at
the site level, of establishing promising forest
and agroforestry management practices.
MATERIALS AND METHODS
DATA COLLECTION
The assessment was based on a global database of biologic and
economic information on forest and agroforest management options.
Information regarding promising practices and initial costs at the
site level within forested nations representing boreal, temperate,
and tropical regions on six continents were collected using
methods described by Dixon et al. (13) and Moulton and Richards
(33). Regional and national biologic and financial data were
collected in three major categories:
1. Forest growth or conservation, as measured by
bioma.ss accretion resulting from forest and
agroforest management practices;
2. Associated costs for each management practice; and,
3. Area of land potentially suitable for each
practice.
Data base methodology and scope were described previously
(13,56). In a global review involving large amounts of technical
5-3
-------
data from this assessment were considered the best available.
When data on land productivity or biomass accretion were outside
the bounds of the technical literature, the values were not used
in the analysis.
FOREST GROWTH AND CARBON STORAGE
r Growth and yield of forests are normally expressed in terms
of volume of stem wood. Thus, it was assumed that one cubic meter
of stem wood is associated with 1.6 cubic meters of whole-tree
biomass, which includes roots, branches, leaves, etc. (32,42,43).
Whole-tree volume was multiplied by the"density (i.e., specific
gravity) of wood for each species to yield whole-tree biomass.
Finally, it was assumed the carbon content of whole-tree biomass
was 50% (8). Although below-ground carbon accretion is
significant in forest systems, only above-ground accretion was
considered in this assessment (5) .
Graham et al. (16) and Schroeder (41) asserted that the
relevant parameter in terms of carbon cycle calculations is the
average amount of carbon on site over an indefinite number of
rotations. If it is assumed that the system is sustainable and
there is no yield reduction in later rotations, the result is the
same as the average amount of carbon on-site over one full
rotation. Because any number of biological, climatic, or social
events could contribute to some level of yield reduction that
cannot be predicted, the approach presented here may represent an
upper bound (1,45). Carbon accretion and storage was calculated
by summing the carbon standing crop for every year in the rotation
and dividing by the rotation length. This approach assumes that,
at or shortly after harvest, all stored carbon returns to the
atmosphere (33).
COSTS OF MANAGEMENT PRACTICES
The relative costs of promising management options used in
this assessment were estimates of direct costs at the site level
for labor, materials, transportation, and the initial
infrastructure (for up to three years) to employ the options
(13,33,56). Scaling of costs (between small and large projects)
was not considered because previous analyses suggest this approach
may be invalid (37,46). The cost of land was not included in the
analysis, because: 1) land cost varies widely around the world; 2)
land values are difficult to establish where land is held in
common by communities or land is government-owned; and 3) no land
market values exist (50).
Financial data are reported in 1990 US dollars. Costs for
any reference year were adjusted based on the inflation and
exchange rates for individual nations according to the
International Financial Statistics (IFS) tables published by the
5-4
-------
International Monetary Fund (25). A nation's inflation rate for
the reference year, as measured by the Consumer Price Index, was
extracted from the IFS tables. The reference year cost was then
converted to a 1990 value and converted back to US dollars at the
1990 exchange rate (13).
Because forests are periodically harvested and replanted
(1,45), the costs of initiating forest management or establishing
plantations are recurring costs. In estimating costs, it is
important to account for these additional investments that will
occur at more or less periodic intervals in the future (10). The
present value of future costs over a 50-year rotation period was
computed for each practice or management option (13,56).
The net interest rate used was 5%. Cost per ton of carbon
was calculated as the present value of all establishment costs
over a 50 year period divided by mean carbon storage. Costs
computed in this manner do not account for any financial benefits
resulting from the initial investment (19,51).
LAND AREA TECHNICALLY SUITABLE
The carbon accretion and storage values were based on a per
unit area basis (e.g., tons carbon per hectare). The technically
suitable land area for each management practice is required to
estimate a total amount of atmospheric carbon removal and storage.
Both land area and carbon storage for different management
practices were classified within nations by ecoregion following
the system devised by Bailey (6). The broadest level of Bailey's
classification was employed. This level, the domain level,
contains four ecoregion subdivisions: boreal, humid temperate,
dry, and humid tropical. A course distinction was recognized
within each ecoregion between lowland and upland zones (e.g., site
5-5
4.
-------
quality). Forest establishment and management practices were
qualitatively assigned to ecoregions for the purpose of tallying
carbon sequestration potential (56).
Estimates of land available on which to implement forest
establishment and management options are based on earlier
assessments by Grainger (17,18), Houghton et al. (22), and Trexler
(49). Within tropical latitudes, the assessment relies primarily
on Advanced Very High Resolution Radiometer (AVHRR) analyses of
land-use patterns by Houghton et al. (22). For the temperate
zones, national inventories of land-use practice and patterns were
consulted (21,33).
STATISTICAL ANALYSIS
Non-parametric statistical analysis techniques (e.g., median
and inter-quartile ranges) were employed to analyze biologic and
economic data collected for various forest management practices
(11). The Wilcoxon 2-Sample Non-Parametric Test (with continuity
correction of 0.5) was used to test significant differences for
each of the comparisons among boreal, tropical, and temperate
median values.
RESULTS
MANAGEMENT OPTIONS
A wide range of promising forest and agroforest management
practices and technologies were identified to promote carbon
sequestering in the terrestrial biosphere (Figure 1). The
assessment analyzed the opportunities to expand forest carbon
pools across boreal, temperate, and tropical latitudes. Based
upon the median values for carbon sequestration in tons of carbon
per hectare, the following are the five most promising
practice-region combinations, from high to low:
Natural rf>genf»rat i nn in trnpiral latitudes (Figure; lei :
Management of humid tropical forests can result in
storage of up to 195 tons of carbon per hectare (tC/ha).
This reflects the great biomass productivity rates of
natural ecosystems in the humid tropics (9). Estimates
of forest growth rates and carbon accretion rates in
tropical forests vary widely.
Affaresfation in the temperal-.e latitudes (Figure lb):
The relatively high median value (120 tC/ha) likely
reflects the growth rates of plantations established on
marginal agricultural lands, which though medium to poor
for agronomic crop productivity, are often quite
suitable for forest plantation growth (1,23).
F.stahli.ghmpnf nf ayrnfnrP.grry in trnpiral latitudes
(Figure lc): If tree and agronomic crops are cultivated
5-6
-------
together, carbon accretion ranges from 60 - 125 tons per
hectare (t/ha). These practices have been employed by
local peoples for centuries (19,30). These moderately
high carbon sequestration values for agroforestry are
encouraging because this practice is also one that will
supply a sustained flow of goods and services to local
populations (Gregerson et al. 1989).
Reforestation in the tropiral latitnries (ricmre 1r) :
The high median value of this practice (65 tC/ha)
supports the assertion that reforestation in the
tropical latitudes has great potential to sequester and
store carbon. Mean annual increments (e.g., up to 60
m3/ha/yr) for eucalyptus and Caribbean pine for rotations
of 20 years or less can rapidly store carbon. These
plantation crops, however, may not always store the
maximum amount of carbon over an extended period because
short rotations limit biomass accumulation (42).
Reforestation in the temperate latitudes (Figure :bl;
At a median value of 56 tC/ha and an inter-quartile
range of 32 - 96 tC/ha, this approach is the fifth
highest on the list of promising practices for carbon
sequestration. Forestation technology and expertise is
well-developed among nations of temperate latitudes.
The lowest median values among management options evaluated
were for silviculture practices (Figure 1). Silvicultural
treatments, such as thinning and fertilization in plantations,
will likely play a role in adapting forests to warmer, drier
climates. Other investigators have reached the same conclusion
(2,32, 44).
Across boreal, temperate and tropical latitudes, biomass
median estimates of the potential to sequester carbon through
establishment and management of forest agroforest systems are 16
tC/ha, 68 tC/ha, and 66 tC/ha, respectively. The Wilcoxon
non-parametric test indicates that the median values for the
temperate and tropical latitudes were significantly greater than
for the boreal and that the temperate and tropical median values
are not significantly different (p £ 0.05).
5-7
-------
COST OF FOREST MANAGEMENT OPTIONS AT THE SITE LEVEL
Initial costs of forest establishment and management are
least expensive in boreal regions. As management intensity
increases in temperate and tropical regions, initial costs per
hectare escalate accordingly (Figure 2). Natural regeneration,
silvicultural treatment, agroforestry, and forestation are the
least expensive practices within tropical latitudes (1,46,56).
Boreal latitudes: For boreal forest systems, natural
regeneration practices and artificial reforestation
could be implemented at a cost of $90-325/ha (Figure
2a). At carbon storage values of approximately 17 t/ha
and 39 t/ha, respectively, the initial cost of carbon
sequestration for the two practices is $5($4-11) and
$8(53-27)/ton (t). Silvicultural treatments are also a
cost-effective means to manage boreal forest systems at
$74/ha. At a sequestration value of 10.5 tC/ha (Figure
1), the initial cost of carbon sequestration ranged from
$5-76/t. Dixon et al. (13) and Allan and Lanly (1) also
reported that forestation and forest management
practices in boreal systems can be sustained and provide
a high rate of return on initial investment. The costs
of forest establishment in Russia, which contains over
50% of the worlds boreal forests, are a major
determinant in calculating global biologic and economic
potential to sequester atmospheric carbon (29) .
lat.ifndfs: Within temperate regions,
reforestation, afforestation, natural regeneration, and
silvicultural practices are the least expensive forest
• management options for sequestering carbon (Figure 2b) .
Artificial reforestation median cost is $350/ha. At a
sequestration value of 56 tC/ha (Figure 2b), carbon is
stored at an initial cost of $6 ($3-29)/t depending on
site conditions, tree species, and management intensity.
Afforestation can store about 120 tC/ha at a cost of
$260/ha or $2($0.22-5)/tC. Natural regeneration can be
implemented inexpensively at less than $10/ha. At 9
tC/ha (Figure 2b), the carbon sequestration cost is less
than $1($0.01-0.43)/t. Intermediate silvicultural
treatments (e.g., thinning and fertilization) enhance
carbon storage in temperate forests at a median cost of
about $350/ha (Figure lb). The initial cost is
$13($3-158)/tC. In temperate latitudes, establishment
of agroforestry systems costs up to $790/ha, and this
practice stores carbon at 34 t/ha (Figure 1) for an
initial cost of $23 ($14-66)/t.
Tropical. latitudes. The wxdest range of costs were
reported for forest management options within tropical
latitudes (Figure 2). Natural regeneration of forests
and establishment of short-rotation fuelwood plantations
and agroforestry systems can all be established for less
5-8
-------
than $500/ha (50 year basis) {Figure 2c) . Reforestation
and agroforestry can sequester carbon at less than
$10($2-26)/t because of high sequestration values, i.e.,
about 100 tC/ha (Figure 1c). Intermediate silvicultural
treatments (e-.g., thinning and fertilization) stimulate
productivity and can sequester carbon at approximately
(SI.50-36)/t at a sequestration value of 59 t/ha (Figure
1c). Therefore, in the tropics, natural regeneration,
agroforestry, reforestation, and silviculture sequester
carbon at median initial costs of less than $10/t.
These initial costs per ton of carbon sequestered
compare favorably to many non-forest options to
sequester or conserve carbon that are $30/t or more
(34,38) .
COST AND YIELD EFFICIENCY OF NATIONAL PROGRAMS
The previous section revealed that establishment of
plantations or agroforestry systems were cost efficient means of
stimulating carbon sequestration compared to other options
(34,38). A comparison of cost and yield efficiency for selected
nations is presented in Figure 3. Costs of carbon sequestration
in forestation programs were highest in Egypt, New Zealand, Zaire,
and Venezuela. In contrast, costs were significantly lower in
Australia, Brazil, China, Congo, Mexico, US, and Russia. The
remaining nations surveyed were intermediate in reforestation
costs. These calculated values for cost and yield efficiency of
national forestation programs do not consider land rental costs,
but are consistent with earlier estimates (2,3,38).
The efficiency and yield of specific practices (artificial
reforestation and afforestation, natural reforestation,
intermediate silvicultural practices, and agroforestry) for
sequestering carbon are presented in more detail for Russia, US,
and Brazil (Figure 4). These data represent a range of economic
options ($0.5-88/tc) to sequester carbon through forest management
in representative boreal, temperate and tropical biomes.
Collectively, these nations represent 30% of the earth's land
area, and implementation of these practices on a large scale could
stimulate significant carbon sequestration. Moulton and Richards
(33) and Swisher (46) observed similar carbon
5-9
-------
sequestration values and cost trends in their assessments of U.S.
and Latin American forest management options, respectively.
A global synthesis of carbon sequestration and initial costs
at the site level is presented in Figure 5a. Total initial cost
rises gradually up to a carbon storage level of about 55 Gt.
Beyond 55 Gt, the total cost begins to escalate at a more rapid
rate. The marginal cost of sequestering carbon in global forest
systems is approximately $lO/t (Figure 5b).
The distribution of technically suitable land among global
ecoregions that would be required to achieve different levels of
carbon storage is presented in Figure 6a. The analysis was
completed by considering the area of land technically suitable for
different practices in each ecoregion and the amount of carbon
that those practices could store (17,22,49). The slope of the
lines is relatively gradual up to 55 Gt indicating that relatively
large increments of carbon can be stored on relatively small
amounts of land. The slope becomes much steeper at 5 5 Gt as
larger increments of land, and therefore higher establishment
costs, are needed to store additional increments of carbon. More
carbon could be stored, but it becomes less cost effective. The
most productive and least expensive lands would likely be placed
under management first (56) .
A global total of approximately 570 million ha "of land would
be required to store 55 Gt of carbon (Figure 6a). Given current
estimates of land suitability and availability, the distribution
would be 190 million ha in the humid tropics, 220 million ha in
dryland ecoregions within tropical and temperate latitudes, and
160 million ha in the humid temperate zones. Utilization of land
in the boreal zone would only be considered at higher levels of
carbon storage. Improved resolution of land availability
estimates in the future could alter these estimates (22) . Figure
6b illustrates the distribution of carbon storage between
ecoregions (6). At the 55 Gt carbon level, 24 Gt would be stored
in the humid tropics, 20 Gt in dryland ecoregions, and 11 Gt in
the humid temperate zones.
Uncertainty is associated with estimates of land areas
technically suitable and socio-economically available for forest
establishment. For example, Trexler (49) estimated that social,
demographic, political, and other factors could result in a 70%
reduction in the available land estimates for tropical Africa and
Asia that were reported by Houghton et al. (22) . A sensitivity
analysis was conducted to determine the possible effects of a 70%
reduction in land available for forest establishment and
management. A linear reduction in available land area evenly
distributed over all nations and ecoregions would result in a
reduction of total carbon storage potential to about 16.5 GtC.
5-10
-------
Similarly, the technical suitability of ecoregions and site
productivity varies widely within a nation or biome (42).
DISCUSSION AND CONCLUSIONS
Past efforts to develop forest establishment and management
cost estimates at the site level for sequestering and conserving
carbon in the terrestrial biosphere have been preliminary
(2,3,13). Site-level (13), regional (39), national (33,38b), and
global (34,35,56) estimates have been calculated. The biologic
and economic opportunity to conserve and sequester carbon in
forest systems appears significant. This assessment suggests:
1. Forest establishment and management practices
(e.g., natural regeneration, reforestation,
afforestation, and agroforestry) can stimulate
accretion of carbon in forest stands of boreal,
temperate and tropical biomes;
2. Forest and agroforest establishment and management
practices can be used to temporarily store carbon
for less than $30/tC, with median values ranging
from $l-8/tC.
3. Technically suitable land can be identified in
boreal, temperate and tropical biomes of the world
to implement forest management practices; and,
4. Potential carbon accretion and storage in forest
systems may total up to 55 Gt over a 50 year
period.
The current assessment of biologic and cost information from
more than 90 nations worldwide represents the first attempt to
develop a bottom-up global analysis of carbon sequestration
potential in forest systems. The forest management practices
identified here can be applied to a wide range of ecosystems in
boreal, temperate, and tropical biomes (21) . However, before
practices can be widely and successfully implemented, more
consideration must be given to the array of possible economic and
socio-political constraints (49,51,55).
From the perspective of forest biomass productivity,
afforestation in the temperate latitudes, agroforestry in the
tropics, and reforestation in both the temperate and tropical
latitudes are among the most promising options. When considering
initial costs in dollars per ton of carbon, attractive options
include natural and artificial reforestation in boreal latitudes;
natural and artificial reforestation, afforestation and
silvicultural practices in the temperate latitudes; and for the
tropics, reforestation and agroforestry systems appear the most
cost efficient (38). The cost estimates are preliminary and do
not reflect benefits associated with goods and services that flow
from forest sector. Moreover, rapidly changing labor costs in
5-11
-------
Russia and developing nations significantly influence economic
analysis of forest management options (19,29,36). The ultimate
mixture of greenhouse gas reduction options for key nations and
the global community (e.g., forest management, alternative fuels,
conservation agriculture) will be driven by the socio-economic and
political factors (4,14,15,19).
The costs of carbon sequestration options at the national
level have been the focus of several recent research efforts in
the United States (33), Germany (52), the Netherlands (401) ,
Brazil (14), Costa Rica (46), and other nations (21). In
addition, analyses of the impact of the Tropical Forest Action
Plan on carbon sequestration have been completed for some nations
and regions (1). The cost estimates of national carbon
sequestration efforts in the current study suggest that programs
could be effectively established in many of the major forested
nations of the world (35,55).
Forest management programs and agroforestry programs have
been implemented in several nations with a range of tree species,
site conditions, and financial support (19,48,51). A number of
the national forest-based programs are for the purpose of carbon
sequestration and conservation. The ability of these programs to
stimulate carbon sequestration varies (21) . It must be stressed,
that a key factor in successful forest management and agroforestry
programs is the involvement and support of local populations in
the planning and implementation phases (19,30,51).
Past and current analyses suggest the next step is micro- and
macro-modeling of the biologic and economic potential of carbon
conservation and sequestration efforts. Regional, national and
global carbon budgets (anthropogenic and biogenic pools and flux)
can be simulated with various process models (5,56). The menu of
forest establishment and management options developed in this
report can be used to define appropriate options to reduce
accumulation of atmospheric greenhouse gases on regional, national
and global scales. Ultimately, modeling efforts should link
forest establishment and management costs with benefits that flow
from the forest sector (e.g., goods and services) (5,33). Such an
approach has been used for preliminary evaluations of forest
sector policy options at national and global levels (2,3,4,38).
REFERENCES
1. Allan, T., and J.P. Lanly. 1991. Overview of status and
trends of world forests. In: D. Howlett and C. Sargent, eds.
op cit. pp 17-39.
2. Andrasko, K. 1990a. Climate Change and Global Forests:
Current Knowledge of Potential Effects, Adaptation and
5-12
-------
Mitigation Options. FO:MISC/90/7. Forestry Department, UN
Food & Agriculture Organization. Rome, Italy. 7 6 pp.
3. Andrasko, K. 1990b. Forestry. In: D. Lashof and D. Tirpak,
eds. Policy Options for Stabilizing Global Climate. USEPA
21P-2003.1. US Environmental Protection Agency. Washington,
D.C. pp. 175-237.
4: Andrasko, K., K. Heaton, and S. Winnett. 1991. Evaluating the
costs and efficiency of options to manage global forests. In:
D. Howlett and C. Sargent, eds. op cit. pp. 216-233.
5. Apps, M.J. and W.A. Kurz. 1991. Assessing the role of
Canadian forests and forest activities in the global carbon
balance. World Resources Review 3:333-344.
6. Bailey, R.G. 1989. Explanatory supplement to ecoregions map
of the continents. Environmental Conservation. 16:307-309.
7. Bouwman, A.F. 1990. Exchange of greenhouse gases between
terrestrial ecosystems and the atmosphere, in: Soils and the
greenhouse effect: Proceedings of the international
conference on soils ana the greenhouse effect. A.F. Bouwman,
ed. Wiley, New York, NY. pp. 62-127.
8. Brown, S. and A.E. Lugo. 1982. The storage and production of
organic matter in tropical forests and their role in the
global carbon cycle. Biotropica. 14:161-187.
9. Brown, S., J.R. Gillespie, and A.E. Lugo. 1989. Biomass
estimation methods for tropical forests with applications to
forest inventory data. Forest Science. 35:881-902.
10. Davis, L.S. and K.M. Johnson. 1987. Forest Management.
McGraw-Hill. New York, NY. 7 90 pp.
11. Devore, J., and R. Peck. 1986. Statistics, The Exploration
and Analysis of Data. West Publishing Company. St. Paul, MN,
699 pp.
12. Dixon, R.K. and D.P. Turner. 1991. The global carbon cycle
and climate change: Responses and feedbacks from below-ground
systems. Environmental Pollution. 73:245-262.
13. Dixon, R.K., J.K. Winjum, and O.N. Krankina. Afforestation
and forest management options and their costs at the site
level. In: D. Howlett and C. Sargent, eds. op cit. pp
319-328.
14. Fearnside, P.M. 1989. Contribution to the greenhouse effect
from deforestation in Brazilian Amazonia. Paper presented to
5-13
-------
the International conference on soils and the Greenhouse
Effect, Wageningen, The Netherlands, 14-18 August, 1989. 43
pp.
15. Goodland, R., E. Asibey, J- Post, and M. Dyson. 1990.
Tropical Moist Forest Hardwoods: The Urgent Transition to
Sustainability. The World Bank. Washington, D.C. 36 pp.
16. Graham, R.L., R.D. Perlack, A.M.G. Prasad, J.W. Ranney, and
D.B. Waddle. 1990. Greenhouse Gas Emissions in Sub-Saharan
Africa. ORNL-6640. US Department of Energy Oak Ridge National
Laboratory, Oak Ridge, TN. 135 pp.
17. Grainger, A. 1990. Strategies to control deforestation and
associated carbon emissions in the humid tropics. In:
Proceedings of the Intergovernmental Panel on Climate Change,
Conference on Tropical Forestry Response Options to Global
Climate Change. US Environmental Protection Agency.
Washington, D.C. pp. 105-119.
18. Grainger, A. 1988. Estimating areas of degraded tropical
lands requiring replenishment of forest cover. International
Tree Crop Journal 5:31-61.
19. Gregerson, H., S. Draper, and D. Elz. 1989. People and Trees:
The role of social forestry in sustainable development. EDI
Seminar Series. World Bank. Washington, D.C. 273 pp.
20. Hill, L.W., H. Heiner and R. Pardo. 1992. Fashioning global
forestry guidelines. Journal of Forestry 90:9-11
21. Howlett D. and C. Sargent, eds. 1991. Technical Workshop to
Explore Options for Global Forestry Management. International
Institute for Environment and Development. London, UK. 34 9
pp.
22. Houghton, R.A., J. Unruh, and P.A. Lefebvre. 1991. Current
land use in the tropics and its potential for sequestering
carbon. Xa: D. Howlett and C. Sargent, eds. op cit. pp.
297-310.
23. Hughes, J.H. 1991. A brief history of forest management in
the American South: Implications for large-scale
reforestation to slow global warming. In: Proceedings of
the International Workshop on Large-Scale Reforestation.
(J.K. Winjum, P.E. Schroeder, and M.J. Kenady, eds.) USEPA,
EPA/600/9-91/014, Washington, D.C., p.53-68.
24. International Monetary Fund (IMF). 1990. International
Financial Statistics Yearbook. Vol XLIII. International
Monetary Fund. Washington, D.C. 775 pp.
5-14
-------
25. International Monetary Fund (IMF). 1991. International
Financial Statistics. Vol XLIV, No 6. International Monetary
Fund. Washington, D.C. 602 pp.
26. Intergovernmental Panel on Climate Change (IPCC). 1990.
Climate Change, The IPCC Scientific Assessment. Working Group
1 Report. WMO and UNEP. University Press, Cambridge, UK. 365
pp.
27. Johnson, M.G. and J.S. Kern. 1991. Sequestering carbon in
soils: A workshop to explore the potential for mitigating
global climate change. EPA/600/3-91/031. USEPA, Environmental
Research Laboratory, Corvallis, OR. 85 pp.
28. Kauppi, P.E., K. Mielikainen,, K. Kuuselo. 1992. Biomass and
carbon budget of European Forest, 1971-1990. Science 256:70-
74.
29. Krankina, O.N. and R.K. Dixon. 1992. Forest management in
Russia: challenges and opportunities in the era of
perestroika. Journal of Forestry. 90:29-34.
30. MacDicken, K.G., and N.T. Vergara. 1990. Agroforestry:
Classification and Management. Wiley, New York, NY. 382 pp.
31. Maini, J.S. 1991. Towards an international instrument on
forests. In: D. Howlett and C. Sargent, eds. op cit. pp.
278-285.
32. Marland, G. 1988. The prospect of solving the C02 problem
through global reforestation. DOE/NBB-0082. U.S. Department
of Energy, Office of Energy Research, Washington, D.C. 66 pp.
33. Moulton, R.J. and K.R. Richards. 1990. Costs of Sequestering
Carbon through Tree Planting and Forest Management in the
United States. General Technical Report WO-58. US Department
of Agriculture Forest Service,.Washington, D.C. 48 pp.
34. National Academy of Science (NAS). 1991. Policy Implications
of Greenhouse Warming. U.S. National Academy Press.
Washington, D.C. 127 pp.
35. Noordwijk Ministerial Conference. 1989. The Noordwijk
Declaration on Climate Change. Ministerial Conference on
Atmospheric Pollution and Climate Change, Noordwijk, The
Netherlands. 15 pp.
36. Peters, C.M., A.H. Gentry, and R.O. Mendelsohn. 1989.
Valuation of an Amazonian rainforest. Nature. 339:655-656.
5-15
-------
37. Row, C. 1978. Economics of track size in timber growing.
Journal of Forestry. 76:576-582.
38. Rubin, E.S., R.N. Cooper, R. A. Frosch, T.M. Lee, G. Marland,
A.H. Rosenfeld, D.D. Stine. 1992. Realistic mitigation
options for global warming. Science 257:148-49, 261-266.
39. Sargent, C. and M. Lowcock. 1991. Options for the
coordination of international action on forest conservation
and management, in: D. Howlett and C. Sargent, eds. op cit.
pp. 163-185.
40. Schotveld, A. 1991. The potential contribution of forestry in
tropical countries to decrease atmospheric carbon content.
in: d. Howlett and C. Sargent, eds. op cit. pp. 291-296.
41. Schroeder, P.E. 1992. Carbon storage potential of short
rotation tropical tree plantations. Forest Ecology and
Management. (In press).
42. Schroeder, P.E. and L. Ladd. 1991. Slowing the increase of
atmospheric carbon dioxide: a biological approach. Climatic
Change. 19:283-2 90
43. Sedjo, R.A. 1989. Forests to offset the greenhouse effect.
Journal of Forestry. 87:12-15.
44. Sedjo, R.A. and A.M. Solomon. 1989. Climate and forests. In:
Greenhouse Warming: Abatement and Adaptation. N.S. Rosenberg,
W.E. Easterling, P.R. Crosson, and J. Dormstadter, eds.
Resources for the Future, Washington, D.C. pp. 105-119.
45. Smith, D.M. 1962. The Practice of Silviculture. Wiley, New
York, NY. 578 pp.
46. Swisher, J.N. 1991. Cost and performance of C02 storage in
forestry projects. In: D. Howlett and C. Sargent, eds. op
cit. pp. 329-339.
47. Tans, P.P., I.Y. Fung, and T. Takahashi. 1990. Observational
constraints on the global atmospheric CO2 budget. Science.
247:1431-1438.
48. Trexler, M.C. 1991a. Estimating tropical biomass futures: A
tentative scenario. In> D. Howlett and C. Sargent, eds. op
cit. pp. 311-318.
49. Trexler, M.C. 1991b. Agroforestry in Guatemala: Mitigating
global warming through social forestry, in: Proceedings of
the International Workshop on Large-Scale Reforestation. J.K.
5-16
-------
Winjum, ?.E. Schroeder, ana M.J. Kenady, eds.
EPA/600/9-91/014. 'JSEPA. Washington, D.C. pp. 37-92.
50. Trexler, M.C. 1991c. Minding the Carbon Store: Weighing U.S.
Forestry Strategies to Slow Global Warning. World Resources
Institute. Washington, D.C. 81 pp.
51. Vincent, J.R. 1992. The tropical timber trade and
sustainable deelopment. Science 256:1651-56.
52. Volz, H.A., W.U. Kriebitzsch, ana T.W. Schneider. 1991.
Assessment of potential, feasibility and costs of forestry
options in the temperate and boreal zones. XD.: D. Howlett
and C. Sargent, eds. op cit. pp. 124-158.
53. Waring, R.H., and W.H. Schlesinger. 1985. Forest
Ecosystems, Concepts and Management. Academic Press,
Orlando, FL., USA, 340 p.
54. White, R. 1991. Technological Challenge at the Earth Summit.
The Bridge. 21(4) ) : 5-11.
55. Winjum, J.K., R.A. Meganck and R.K. Dixon. 1991b. Toward a
global forest agreement: opportunities to expand and manage
world forest systems. Journal of Forestry, in press.
56. Winjum, J.K., R.K. Dixon and P.E. Schroeder. 1992.
Estimating the global potential of forest and agroforest
management practices to sequester carbon. Water, Air and
Soil Pollution 64:213-228.
ACKKOWLEDGMENTS
The authors thank P.M. Bradley, G.A. Baumgardner and S.G.
McCannell for their technical assistance in preparing this
document. This manuscript was prepared from a chapter in the EPA
ORD report
600/3-91/067 "Assessment of Promising Forest Management Practices
ana Technologies for Enhancing the Conservation and Sequestration
of Atmospheric Carbon and Their Costs at the Site Level".
5-17
-------
Figure 1 Carbon storage tor forest management options for (a) boreal, (b) temoerate, and (c) tropical
biomes. Meaian values are indicated by the wide honzontal lines. Boxes represent
interchange ranges (middle £0% of observationsshadea circles are means ana vertical
lines indicate full range of data.
(a)! ' ' ~~ "~
7C :
so
50
s *0 J
— 3C «
20 i
<
j
10 j
r,i\
n)
{«>
!19)
•:i2)
(101
m
Reforasia'jcn A"cresiai.c^
.'Jaajraj
regeneraaon
Snort
rata: on
Boreal
Management option
(b)i
3S0 i
200
250
2
200
c_>
1S0
iao
so
0
'212)
(119)
(6)
(i)
(62)
(10)
(2)
eh
¦s
-frr
Temperate
Rcfortsaaon Attoruoaon Nuni SiMcuftur* Short roaacn Agrsforestry
rvgtnmoon
Management option
(C)
700
600
soo
o
400
j=
u
300
200
100
0
(136)
(3)
(«>
(3)
(12)
(16)
nr
•Raiorasaaon Afloraaaoon
Na&M
mfntiMon
SMajmjra
Agrolorasiry
Tropical
Management option
_ i*
5-18
-------
Figure 2 initial costs for forest management options for (aj boreal, (b) temperate, and (c) tropical
biomes. Median values are indicated by the wide horizontal lines. Boxes represent
interquartile ranges (middle 50% of observations), shaded circles are means and vertical
lines indicate full range of data.
600
soo
4
4
MO 'i
300
200
ISO
(16)
(1)
(n)
(19)
(13)
Boreal
-&
Raforasadon
A Wort ran on
Natural
faganarason
SlivicutUr*
Sfton
maoon
Management option
(b)
(n)
(80)
(64)
(10)
[11)
*000
3500 ¦
3000 -<
2500 •
1500 ¦
1000 ¦
500 ¦
Raforasaoon Afloraiason
Nauru
SBvtcuRur* Snort rotation Agreforasiry
Temperate
raganaradon
Management option
(C)
«S00
4000
3500
3000
^ 2500
^ 2000
1500
1000
500
0
(75)
£3
Tropical
(2)
(3)
(3)
m
Raforasaoon Aiforasadon
Management option
NauaJ-
wganafaoan
SOvtcutoir*
(13)
Agroforaspy
5-19
-------
Figure 3 Initial cost and carbon storage efficiency of national reforestation programs.
so
~€GY
70
60 -
50
~VEN
~NZL
"5 40 -
30
20 —
10 -r
~SUN
~ZAR
DEU 4CIV
'CRI
FIN
~CAN
~SRA„
~ARG
~COL
ECU
~
~PHI
CHL
-tWA
~FRA
~HVO
us. ~°NJ5f
}M^0=
XfMYS
PAK.
IV
20 40 60 SO 100 120
Carbon Storage (IC / ha)
« GHA
~SEN
h
~TGO
140
160
180
Code
Nation
tC/ha
S/tC
Code
Nation
tC/ha
S/tC
Code
Nation
tC/ha
S/!C
ARG
Argentina
65.0
25.0
EGY
Egypt
36.5
77.9
PAK
Pakistan
98.8
2.5
AUS
Australia
107.0
5.9
FIN
Finland
40.9
27.3
PHL
Philippines
44.6
2.7
BRA
Brazil
63.5
27.4
FRA
France
115.9
16.6
ZAF
S. Africa
110.0
8.7
HVO
Burkina Fasc
92.8
13.3
DEU
Germany
56.0
32.4
SEN
Senegal
128.8
2.9
CAN
Canada
40.7
11.5
GHA
Ghana
129.9
6.0
THA
Thailand
79.2
1.7
CHL
Chile
75.9
2.7
IND
India
78.0
26.8
TGO
Togo
167.0
4.6
CHN
China
76.3
5.2
IDN
Indonesia
99.0
6.8
USA
USA
77.0
5.5
COL
Colombia
56.7
20.6
MDG
Madagascar
75.0
1.8
SUN
USSR
25.0
4.6
COG
Congo
111.0
3.0
MYS
Malaysia
84.0
3.5
VEN
Venezuela
38.9
50.7
CRI
Costa Rica
46.3
31.0
MEX
Mexico
99.0
4.5
ZAR
Zaire
59.5
41.7
CIV
Cote cflvoire
60.2
34.1
NZL
N. Zealand
94.0
52.4
ECU
Ecuador
67.2
8.7
NPL
Neoal
78.6
0.5
5-20
-------
Figure 4 Efficiency and yield of reforestation, afforestation, natural regeneration,
agroforestry and silvicutturai practices in Russia, US, and Brazii.
30
r o
' 25
20
o
15
10
o
r
O
5
0
Russia
5/tC = 15.5 - 0.4S(tC/ha)
r=0.43
"» I
10
Total tons Carbon / hi
15
20
o
US
S/tC = llea5/"OU)
r = 0"8
jL,
100 150 200 250 300
Toui tons Carbon / ha
Legend
O Ralorastation
^ AHoraatatien
| Natural
flagtneratlon
^ Silviculture
X Agroloreslry
100
so
a 60
— 40
20
0
x
o a
V_i
Brazil
S/lC=13.4e»/,ow
1^=0.40
.O ^
CP ~
50
100 150
Total tona Carbon / ha
200
250
5-21
-------
Figure 5a Total initiaJ global cast of sequestering cartoon in forest systems
employing forestation and forest management practices.
500
450
ST 400
^ 350
mt
— 300
1 250
:| 200
5 150
5 100
50
0 te
10 20 30 40 50
Toui ion* C (z 10*9)
SO
70
80
Figure 5b Marginal initial costs of sequestering carton in forest systems
employing forestation and forest management practices.
120
100
so
"5
|
60
3
40
I
20
0
10 20 30 40 50 60
Carbon aiervd (tons *10*9)
70
80
5-22
-------
Figure 6a Potential distribution of technically suitable land among ecoregions
of key forest nations worldwide for different levels of carbon
storage.
40 so
Tan* C (Gl)
Hurwd
TrooicaJ
Dry
Tropical
Humid
Temperas
BoreaJ
30
90
Figure 6b Potential distribution of stored carbon among ecoregions of key
lorest nations worldwide for different levels of total carbon storage.
60
50
g 40 ¦¦
30 ¦¦
20 ¦¦
10 ¦¦
0
Humid
Tropical
Dry
Tropical
. Humid
Temperate
Boreal
0.6 0.8 1 \2 1.4
Tola! hectares (bllllona)
1.6
1.8
*
5-23
-------
GLOBAL BIOME
(BlOspheric Mitigation and adaptation Evaluation)
PROGRAM
Program Description
U.S. Environmental Protection Agency
Office of Research and Development
Robert K. Dixon and Jack Winjum
U.S. Environmental Protection Agency
Corvallis, OR
(This paper has been reviewed in accordance with the U. S. ttOST4
Environmental Protection Agency's peer and adminis-
trative review policies and approved for presentation ^ *£
and publication.) ^ J?
%
ORD/ERL-C
5-24
-------
Preliminary assessments suggest that forests and agroecosystems can be managed to conserve and
sequester carbon, thereby reducing the accumulation of greenhouse gases in the atmosphere. Biomass
utilization is a necessary component of a sustained terrestrial carbon sequestration effort. The Global
BIOME (BlOspheric Mitigation and adaptation Evaluation) Program will consist of: a) technical
assessments of effectiveness of terrestrial biosphere management options and biomass fuel technology
in reducing atmospheric accumulation of greenhouse gases, b) demonstration projects to assess the
technical and economic feasibility of applying agricultural and forest management options and biomass
fuel substitution, c) regional, national, and global assessments of effectiveness of terrestrial biosphere
management and adaptation options, and d) assessment of practices and technologies which, if
implemented, could facilitate adjustment of forest and agroecosystems to global climate change. The
Global BIOME initiative is a component of EPA's Office of Research and Development national Global
Change Research Program. The research is managed by the Agency's laboratories: ERL-A, ERL-C,
AEERL, and AREAL. Research is conducted by EPA scientists in cooperation with universities, other
federal agencies and laboratories, and contractors.
PROGRAM GOAL
Evaluate the degree to which forest and agro-
ecosystems can be technically managed or adapted,
on a sustainable basis, to conserve and sequester
carbon, and the degree to which biomass may be
substituted for fossil fuels to reduce the accumu-
lation of greenhouse gases in the atmosphere.
Emphasis will be placed on managed terrestrial
ecosystems given their significant role in the global
carbon cycle. Appropriate biologic management
and adaptation technologies, costs and benefits,
implementation procedures and environmental risks
and benefits will be assessed.
POTENTIAL BIOLOGICAL GLOBAL CARBON
(mbovo- mnd boiow-ground)
Ml TIG A TION
Seau6sti*ation
U.S.
Tmmpmrmtm Tropic*!
Boreal
Foreatatlon
O.IO
0.8
I.O
0.1
Agroforeatry
0.05
0.1
0.5
Revegetatlon
0.05
0.2
0.2
0.2
Silviculture
0.03
0.1
0.1
0.1
Conservation
Reduce deforestation
- . ¦
1.S
0.1
Halt cJeaertlflcatlon
0.2
0.2
Fire management
0.2
-
0.2
Total
0.23
1.6
3.5
0.7
OlmMt Niatmis (im), (Mian,
IM
•4. <1M1>
n>
A range of terrestrial biosphere management options which conserve or sequester carbon are available for
utilization in the US. and biomes ¦worldwide. Based on current estimates application of forest and agricultural
management practices on global scale could potem tally sequester and/or conserve I -6 CiC annually. Many of these
options have value added benefits beyond the reduction of greenhouse gases.
5-25
-------
BACKGROUND
The accumulation of greenhouse gases in the atmo-
sphere due to anthropogenic activities (e.g., defor-
estation, fossil fuel combustion) may have begun
to change the global climate. Given our current
understanding of global carbon sources and sinks,
the prospects for managing the terrestrial bio-
sphere to alter the carbon cycle to mitigate and
adapt to climate change appear promising.
The Global BIOME Program was developed to
answer the following policy question:
Can increases in the concentration
of atmospheric C02 be effectively
reduced oradapted to by increasing
carbon sinks in forest and agricul-
tural systems, and by the substitu-
tion of biomass for fossil fuels?
Sequestration: Forests and agroecosystems play
a prominent role in the global carbon cycle. For-
ests alone contain an estimated 66% of terrestrial
aboveground carbon, and approximately 45% of
the terrestrial soil carbon. Global forests account
for some 90% (90 Gt) of the annually carbon flux
to and from the atmosphere from terrestrial sys-
tems. Based on current estimates, application of
forest and agroecosystem management practices
on a global scale have the biological potential to
sequester or conserve 1-6 GtC annually.
Conservation: Conservation efforts can be em-
ployed to retain carbon in the terrestrial biosphere.
Slowing global deforestation could conserve over
one GtC annually. Agroecosystems contain about
12% of the world's terrestrial soil carbon and
conservation of this pool is essential to sustained
crop productivity and decreasing C02 emissions.
Some agricultural practices have been shown to
increase soil carbon content by increasing carbon
sequestration and/or reducing the loss of carbon.
Examples include reduced tillage, crop residue
incorporation, field application of manure and
sludge, and rotations using cover crops or legumi-
nous crops. Additional benefits from implement-
ing forestry and agricultural practices that con-
Offssts of us Carbon Emissions Dy Forestry Practices
Coogtrvatioi Reserve
iicreaseo P'oojetivuy
Uro®-> Trees
Afforestation
Bismais Energy
Total
Year 2000
vear 2015
• y/s'///
O 2 * 6 8
Potential orfset aa a percentage of 1987 emissions
Off ice of Teermology Assessment 1991
Forest management and biomass utilization practices can be applied in the U.S. to conserve and
sequester carbon and offset Commissions. Establishment of trees in marginal lands, expansion of
forest plantation, urban tree planting, and substitution of biomass for fossil fuel are all viable near-
term options to slow accumulation of greenhouse gases in the atmosphere.
5-26
-------
serve soi 1 carbon include increased soil water hold-
ing capacity, and nutrient availability, improved
soil physical properties, and decreased soil erosion
by wind and water. The Global BIOME Program
will assess the forest and agroecosystem contri-
bution to global carbon conservation.
Fuel Substitution: The potential exists to substi-
tute biomass energy for some portion of fossil fuels
to reduce the most significant anthropogenic factor
in global warming, i.e., atmospheric additional
cases, of C02 from fossil fuel burning. The use of
biomass would also encourage the establishment
of agricultural and forest systems which will cap-
ture and sequester C03. Utilization of biomass is
a requisite for a significant carbon sequestration
program since forestry and agricultural manage-
ment options are finite and stored terrestrial carbon
can eventually produce greenhouse gases. Cur-
rently, harvesting is responsible for 40% of total
biomass costs. Preparation of biomass for various
combustion processes may also be expensive.
Technological advances in both of these areas
should substantially reduce the cost burden and
must be addressed.
Adaptation; The projected rate and magnitude of
climate change impacts on forest and
agroecosystems is unprecedented in recent geologic
history. Adaptation of terrestrial ecosystems to
climate change will be necessary to maintain a flow
of goods and services, as well as to conserve and
sequester carbon. Biologic options to manage
forest ecosystems to facilitate ecosystem adaptation
to climate change, including forestation, thinning,
and genetic manipulation will be evaluated.
Similarly, adaptation of agroecosystems may be
achieved through alternative cropping practices or
land-use systems.
Recently, the G-7 countries, recognizing the
prominent role of forest biomes in global ecology
and the global carbon, agreed to develop a conven-
tion process to promulgate a Global Forest
Agreement. The stated intent of this agreement
includes:
• curb deforestation,
• protect biodiversity,
• stimulate sustained forest management
and productivity, and
• address threats to the world's forests.
The Global BIOME Program will support this
process by assessing the current distribution and
extent of manageable forests, the suitability and
availability of land for forest ecosystem manage-
ment, and the state of practices which can be
applied to forest management on a global basis.
PROGRAM ELEMENTS
Each Program Element and the scope of work to
complete component tasks are presented below.
ELEMENTT: Regional and National Assessment
of Effectiveness of Terrestrial Biosphere Manage-
ment Options
Rationale: Evidence suggests the terrestrial bio-
sphere can be managed to conserve and sequester
carbon and facilitate adaptation to global climate
change. An assessment of the capability to con-
serve and sequester terrestrial carbon requires
knowledge of land suitability and availability, ter-
restrial systems management and adaptation, op-
tions, biomass utilization options, and ecoregional
risk and benefits, in key regions and nations.
Major program outputs: Completion of tasks in
Element I will provide a summary of options, and
identify target/key countries for pilot demonstra-
tions.
Approach: Five major tasks will be completed in
the key country assessment:
A. Carbon Pools and Flux: Carbon budgets and
models will quantify pools and flux for specific
regions and nations. A detailed carbon budget for
5-27
-------
the key countries, including biogenic and anthro-
pogenic components will be constructed based on
current data bases. These carbon budgets will be
assembled by aggregating data for specific pools.
Emphasis will be placed on biogenic components.
Simulation models of anthropogenic and biogenic
carbon pool dynamics will be developed. The
models will be employed to evaluate effectiveness
of various mitigation/adaptation strategies (e.g.,
tree planting versus emission reduction).
B. Land Suitability and Availability Character-
ization: Land-use practices and land availability
will partially define opportunities to implement
tenestrial biosphere management options. This
assessment will assemble and evaluate all data on
land technical suitability and availability for key
nations. Technical suitability is defined as having
edaphic, climatic, and biological conditions that
permit the management of tree and/or agricultural
crops on a sustainable basis for the purpose of
conserving/sequestering carbon. Availability is
based upon the social, political, and economic
limitations of the countries with technically suit-
able lands. Further, an assessment of land-use
practices and projected climate change on land
suitability and availability is required.
C. Forests and Agroecosysterns' Adaptation
Options: Research will assess how shifts in forest
distribution and productivity influence species
survival and which species are most resilient in the
face of altered climatic conditions. Management
strategies and silvicultural practices will be iden-
tified which allow forest systems to adapt to warmer
and drier environments. Options to be investigated
include both in situ (e.g., nature reserves, migration
corridors) and ex situ (e.g., zoos, botanical gar-
dens) measures. Appropriate technology and man-
agement options will be ranked by: 1) carbon
conservation and sequestration efficiency; 2) op-
erational feasibility; and 3) socio-economic and
eco-risk factors.
D. Biomass Utilization Options: Substitution of
biomass for fossil fuels offers significant opportu-
nity to reduce emissions of greenhouse gases. This
analysis will assemble and evaluate available data
to rank the technological and economic feasibility
of utilizing cultivated biomass from forestry and
agricultural systems as a substitute for fossil fuels.
The utilization of biomass will be viewed as a
potential source of renewable energy, as feedstock
for conversion to substitute fuels for transporta-
tion, or for use in durable products. This research
will assess: 1) supplies of biomass from agricul-
tural and forest residues; 2) the market potential for
biomass energy technologies in the presence of
alternative fuels; 3) biomass harvesting options;
and 4) environmental and energy impacts of bio-
mass combustion (conventional and advanced pro-
cesses).
E. Risk-Benefit Assessments: The application of
terrestrial biosphere management options will be
dependent on social and environmental risks and
benefits. Most of the management options to
sequester and/or conserve carbon have collateral
value-added benefits to society such as production
of commodities, protection of biodiversity, and
employment ami income opportunities. A model
framework will be developed for the examination
Agrcforestry is a combination of trees and agronomic
crops. A wide array of agroforestry systems are
established and managed within temperate and tropical
latitudes to produce food, fuel, and fiber. The net
primary proauca vity of these systems is relatively high
and the potential to sequester and conserve carbon is
approximately 2 Ct, globally.
5-28
-------
of trade-offs between: 1) ecological risks; 2) social
and economic cost/benefits and effectiveness of
decreasing atmospheric greenhouse gas emissions.
Field assessments, literature reviews and model
simulations will provide an assessment of the po-
tential feasibility and benefits of terrestrial carbon
adaptation/mitigation measures. The country-spe-
cific data collection, and analysis of the availability
and technical suitability of lands for the sequester-
ing of carbon in biodc systems requires careful
scrutiny of management opportunities, socio-eco-
nomic costs and benefits, and potential eco-risk.
ELEMENT Tf: Pilot Projects to Assess Terrestrial
Biosphere Management Options
Rationale: Demonstration trials must be estab-
lished in key regions and nations to determine
effectiveness of terrestrial biosphere management
options.
Major program output: Technological feasibil-
ity of management options and costs/benefits at
site level will be characterized.
Approach: Pilot projects and demonstration trials
will be established in key nations (e.g., U.S.A.,
Brazil, Mexico, Malaysia, Russia) to assess
effectiveness of terrestrial biosphere management
and adaptation options. The pilot projects will
evaluate biosphere management technologies and
logistics, including biomass utilization, in specific
ecoregions/biomes. Research will provide
defensible data and models for quantifying the
carbon pools and dynamics for forestry and
agricultural systems, critical soil and climate
parameters, and the effects of management and
adaptation approaches to conserve and sequester
carbon. An assessment of social, economic and
ecological risks and benefits will also be conducted
at these field sites. The pilot projects will be long-
term, intensive efforts to provide data for carbon
budget and model development and calibration.
Demonstration trials will be employed for
technology transfer to various clients.
The pilot projects and demonstration trials will be
established at extramural sites. Co-location and
cooperation with U.S. Agency for International
Development, U.S. Department of Energy, U.S.
Department of Agriculture and others will be en-
couraged in the U.S.A. and abroad. In-country
support will be a key to large-scale international
pilot projects and demonstration trials.
SEQUESTRATION OF
CARBON IN FORESTS
Forests occupy approximately one-third of the Earth's land area and their vegetation and soils contain about 60%
of the terrestrial carbon. The Earth's atmospheric CO, cycles through the terrestrial biota every seven years, and
of this amount 70% cycles through forest ecosystems. These features suggest forest and agroecosysiem
management practices can be employed to mitigate the accumulation of greenhouse gases, such as C03. in the
atmosphere. Carbon sequestration rates in forest and agroforest systems range from 1-8 tC/ha/yr.
5-29
-------
ELEMENT TTT: Global Assessment of Teiicstrial
Biosphere Management
Rationale: Preliminary assessments suggest the
global application of terrestrial biosphere manage-
ment options have the potential to reduce the ac-
cumulation of greenhouse gases in the atmosphere
and mitigate global change processes. The global
effectiveness of these options will need to be
evaluated by integration into large-scale models
(e.g., GCMs or ESMs).
Major Program Output: A synthesis of the ef-
fectiveness of terrestrial biosphere management
options to reduce greenhouse gas emissions to the
atmosphere and mitigation of global climate change.
Approach: A synthesis of knowledge regarding
the current potential for global terrestrial carbon
conservation and sequestration will be provided.
Carbon budget and modeling output from key
national assessments and other data sources will be
synthesized and evaluated on a global scale. An
array of management and adaptation options (e.g.,
forestation, substitution of biomass for fossil fuels)
will be examined. The global assessment will
include an evaluation and synthesis of anthropo-
genic and biogenic carbon pools and flux world-
wide.
Biogeochemical and Earth System models of global
carbon pools and flux will be employed to project
future carbon flux rates to the atmosphere. The
effectiveness of terrestrial biosphere management
options to reduce carbon flux to the atmosphere
will be examined. General circulation models,
coupled with atmosphere-biosphere models, will
be used to consider the responses and feedbacks of
the terrestrial-biosphere to project future climate
change, given the application of carbon cycle
management options.
The global synthesis will be completed in con-
junction with complementary research activities in
the U.S.A. and abroad. Cooperation with existing
and emerging biogeochemical model and GCM
programs will be essential. Econometric, social
and environmen tal decision models forrisk-beneflt
assessment of global management options will
also be employed.
TERRESTRIAL BIOSPHERE
The accumulation of greenhouse gases. a.g. C02. CH4. CO. In the atmosphere is a major perturbation of tha
global cartcn cycle. Tha nat increase in atmospheric C02 is tha result of graatar release than its removal by the
terrestrial biosphere and marine systems. Two sources of C02 are especially significant, the combustion of fossil
fuels and deforestation. The flux of carbon through the terrestrial biosphere, by plant photosynthesis, respiration,
and decomposition is approximately 100 Gt annually. Oceans have large pools of global carbon, but annual nat
flux with the atmosphere is approximately equal to thm of tha terrestrial biosphere.
5-30
-------
REPRESENTATIVE PROGRAM OUTPUTS
Bamweil, T.O.. R.B. Jackson, IV, E.T. Elliort, LG Burke, C.V. Cole. K. Paustaian, E.A. Paul, A.S.
Domgian, A.S. Patwaidhan. A. Roweil and K. Weiniich. An approach to assessment of management
impacts on agricultural carbon. Joumai of Water. Air & Soil Pollution (in press).
Dixon, R.K. and DP. Turner. 1991. The global carbon cycle and climate change: responses and
feedbacks from belowground systems. Environmental Pollution 72:245-262.
Jackson, R.B. Potential for CH4 from animal waste to reduce U.S. fossil fuel CO2 emissions. Climatic
Change (in press).
Lee, J J. and D.A. Lammen. 1990. An approach to the regional evaluation of the responses of soils to
global climate change. Pages 395-399 in: l.Bouman(ed.), Soils and the Greenhouse Effect. John Wiley
and Sons, New York.
Peer, R.L., D.L. Campbell and W.G. Hohenstein. 1991. Global warming miiigaiion potential of three
tree plantation scenarios. EPA-600/7-91-003. U.S. Environmental Protection Agency. Research
Triangle Park, NC.
Schroeder, P.E. 1991. Can intensive management increase carbon storage in forests? Environmental
Management 15:475-481.
Smith, K.R., R.A. Rasmussen, F. Manegdeg and M. Apte. 1992. Greenhouse gases from small-scale
combustion in developing countries. EPA-600-R-92-005. U.S. Environmental Protection Agency,
Research Triangle Park. NC.
Winjum, J.K., RJC Dixon and P.E. Schroeder. 1992. An assessment of forest management practices
for sequestration of carbon: estimating the global potential. Journal of World Forest Resource
Management (in press).
Winjum, J.K., R.A. Meganck and RX Dixon. 1992. Toward a global forest agreement: Opportunities
to expand and manage world forest systems. Journal of Forestry (in press).
For more information contact:
R. Dixon T. Barnwell
ERL-C ERL-A
503-754-4600 404-546-3134
5-31
-------
AGRICULTURAL MANAGEMENT AND SOIL CARBON SEQUESTRATION:
AN OVERVIEW OF MODELING RESEARCH
Robert B. Jackson, IV, Thomas 0. Barnwell, Jr.
U.S. EPA, Athens, GA 30605 USA
Anthony S. Donigian, Jr., Avinasb S. Patwardhan
Aqua Terra Consultants, Mountain View, CA 94043 USA
Kevin B. Weinrich, Allen L. Rowell
Computer Sciences Corporation, Athens, GA 30605 USA
Abstract. Soil carbon fluxes and pools are profoundly affected by agricultural management, which in
rum is affected by national and international agricultural policies. This paper briefly describes the
framework and some details of the soil and climate databases used in a computer modeling research
strategy that is being implemented at the Environmental Research Laboratory, Athens, Ga. (AERL). The
objective of the research is to determine the potential for U.S. agroecosystems to accumulate and
sequester carbon as a means of slowing the global increase in atmospheric carbon dioxide.
INTRODUCTION
The conversion of native ecosystems to agricultural lands in the central United States was perhaps the
most extensive ecological disturbance known in North America during the past ISO years [Barnwell et
al. 1991, Wilson 1978}. A major consequence of this land conversion has been a substantial loss of
organic matter from soils across the regions [Haas et al. 1957, Campbell 1978, Barnwell et al., 1991].
Globally, soil C losses from land use change and agricultural management have been a primary
anthropogenic source of atmospheric carbon dioxide (CO}), even exceeding fossil fuel CO, emissions until
about SO years ago (Figure 1) [Houghton et al. 1983]. Since soil carbon levels in most Great Plains and
Corn Belt agricultural soils have been reduced to levels well below those existing prior to the
establishment of agriculture, the potential should exist to increase them, thereby providing a sink for
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication.
5-32
-------
6
cr> s
FOSSIL FUELS
r-1
4
BIOTA AND 3OILS
1
0
1860
1900
1920
1940
1960
1980
YEAR
Figure 1 Historical Terrestrial Carbon Emissions [Houghton et al., 1983]
(Reproduced with permission.)
atmospheric C02. While pressures on agricultural management in the United States have been increasing
in response to the environmental and economic costs of energy and chemicals (among others), many of
the techniques designed to address these issues have also demonstrated a potential for increasing soil C
levels above those maintained under conventional agricultural production systems. Such practices include
increased use of green manure and animal manure [Vitosh et al. 1973], reduction in the proportion of
bare fallow relative to crop in semiarid regions [Wood et al. 1991], and reduced or no-till management
[Blevins et al. 1983, Dick 1983, Lamb et al. 1985]. However, the extent to which soil C levels can be
regulated through the application of various combinations of these practices for different cropping
systems, soil types, and climatic regions is not yet clear, particularly since agricultural management is
driven by economics, which in turn, is greatly impacted by national and international agricultural policy.
5-33
-------
To date there has been little systematic evaluation of how policies that encourage or discourage particular
land uses or agricultural management practices affect regional soil C balances, and of what effect future
policy shifts may have.
Data available from long term agricultural research sites have shown that the potential exists to
increase agricultural soil carbon through management alone [Jenkinson et al. 1977 and 1987, Van der
Linden et al. 1987]. The U.S. EPA is interested therefore, in assessing not only the agricultural
management effects on soil carbon, but the policy implications driving the use of alternative agricultural
management systems. To that end, we have developed a modeling framework that begins with the
policies to be evaluated and their effects on agricultural economics; then includes the economy's
subsequent effects on agricultural management through product demand, planted area and crop profit
potential; and finally examines the effects of the resulting agricultural management scenarios on soil
carbon (C) and nitrogen (N> fluxes and pools (Figure 2).
This year's research effort by AERL will analyze the central region of the United States (Figure 3).
National and international agricultural policy scenarios are generated by EPA's Office for Planning and
Policy Evaluation (OPPE). These policies then become inputs to the Comprehensive Environmental and
Economic Policy Evaluation System (CEEPES) model known as RAMS [CARD 1992], RAMS is an
Flux+a and Pools |
Figure 2 Modeling Framework for Assessing The Impact of Policy on
Agricultural Management, and Agricultural Managements' Impacts on Soil C
Fluxes.
5-34
-------
Figure 3 First Year Study Region Displaying Production
Areas in bold (PA) and State Boundaries (Note: PAs follow
county boundaries).
economic model maintained at the Center for Agricultural and Rural Development (CARD) at Iowa State
University. The RAMS model divides the United States into production areas (PA) (Figure 3) for which
crop areas and rotations are determined based initially on policy scenarios. CARD has tentatively
specified twenty-seven PA's in our study area (Figure 3).
Twelve soil C models were reviewed before two were selected for our analysis [Barnwell et al.
1991]: the CENTURY model [Parton et al. 1983, Parton et al. 1987, Parton et al., 1988; Parton et. al.
1989] and the Denitrification Decomposition model (DNDC) [Li et al. 1992, Li et al. 1992a & b].
Among the numerous factors influencing soil C and N fluxes in both models are initial soil organic matter
and nitrogen levels, climate (precipitation and temperature), crop type and yield, crop rotations, soil
texture. Ullage type and frequency, grazing, fertilizer applications, residue returns or removals, and
irrigation frequencies and amounts. Both of the soil carbon models are site specific, which requires that
for our purposes we develop a methodology for making regional assessments, in addition to performing
model validations using site data.
5-35
-------
SITE DATA FOR VALIDATION
A workshop was held to assess the avaiJable databases for studying C and N fluxes and pools in
North American agricultural soils [Barnwell et al. 1991]. The workshop helped to identify site field data
for model testing, calibration and further validation. The "Soil Organic Matter (SOM) In Temperate
Agroecosystems" workshop was held on February 18-21 1992 in Hickory Corners, Michigan.
Presentations at the workshop included SOM data, as they relate to agricultural management, for more
than SO agricultural research sites in the United States and Canada. These data are currendy being
compiled into a conference proceedings and database [Paul and Elliot 1993], We chose two sites for
calibration and validation of CENTURY. These sites represented two of the primary crops grown in our
study region - corn and wheat.
REGIONAL DATA FOR MODEUNG
Soils.
CENTURY and DNDC require input data on soil texture (percent sand, silt, and clay) and on initial
soil C and N levels. County resolution information was acquired for all of the soil data. These data were
taken from three soils databases - Data Base Analyzer and Parameter Estimator (DBAPE) [Imhoff et al.
1989], the 1982 National Resources Inventory (NRI) [Goebel and Dorsch 1982], and the 1987 NRI /
Soils-5 database [Goebel 1987]. The soil texture information taken from the 1982 NRI Database was
contained in the 1982 NRI / Soils-5 identification block (fields 76-78) in the 1982 NRI file. The file
contained a code with which a texture could be determined from the National Soils Handbook [Soil
Conservation Service 1983]. Both DBAPE and the 1987 NRI databases contained ranges of percent sand
and clay. The high and low values from these ranges were averaged and their sum subtracted from 100
to obtain percent silt. This was done for each layer of soil for every soil type in the counties of the study
region identified in DBAPE and the 1982 NRI databases. The soil texture values were averaged using
depth weighting of percent sand, silt, or clay to a 20-cm depth from the surface of the soil.
-------
The three databases were utilized independently, meaning that for any given county, only DBAPE
data were used (Figure 4), or 1982 NRI data were used (Figure 5), or 1987 NRI / Soils-5 data were used
(Figure 6). Thus combining the files entailed only the sorting of county federal information processing
(fips) codes. The textures in the combined file were then reclassed into 1 of the 12 texture categories
given in the Soil Conservation Service texture triangle. The forma: of the result was " County FIPS,
Texture, Area " for all of the counties in the study region (i.e., each county could potentially have from
1 to 12 different soil textures). Finally, the soils were grouped into six texture classes based on the extent
of different soil textures in the study region (Table 1).
h
r
Figure 4 Counties for Which Soil Textures Were Taken From DBAPE.
5-37
-------
Figure 5 Counties for Which Soil Textures Were Taken From 1982 NRI.
Figure 6 Counties for Which Soil Textures Were Taken From 1987 NRI.
5-38
-------
Table 1 Texture Categories.
USDA Texture Categories Combined Categories"
SAND
LOAMY SAND
SAND / LOAMY_SAND
SANDY LOAM
SANDY CLAY
LOAM
SANDY LOAM / SANDY
_CLAY
_LOAM
LOAM
LOAM
SILTY LOAM
SILTY_LOAM
SILTY CLAY
SILTY CLAY
LOAM
SILTY_CLAY / SILTY_
_CLAY
_LOAM
CLAY
CLAY LOAM
CLAY / CLAY_LOAM
Note: SANDY_CLAY and SILT were not found in the study area.
The models use monthly values for maximum and minimum temperature (CENTURY) or
average temperature (DNDC), and total precipitation. These values can either be provided to
the models as historically-recorded input from a user-supplied file, or the models can
stochastically generate weather using mean and skewness values computed from a user-supplied
data file or input by the user. A collection of historical climate data for the United States
[Wallis et al. 1991] was used to obtain 41 years (1948-1988) of monthly maximum and minimum
temperatures plus total precipitation data for 589 climate stations in the central United States
(Figure 7). From these data, an average monthly maximum temperature value, and an average
annual precipitation value was derived for each of the stations. A Geographic Information
System (GIS) then was used to create contour maps of temperature (Figure 8) and precipitation
(Figure 9). The temperature contours (2° C intervals) and precipitation contours (10-cm
intervals) were then overlayed to produce areas of similar climate (i.e. Climate Divisions [CD]).
The Climate Division (CD) boundaries were adjusted to coincide with county boundaries;
primarily to simplify allocation to the CDs of crop and rotation area data from the RAMS model
and the county level soils data.
5-39
-------
Although this approach successfully identified subregions of like climate, some of the
regions were not contiguous and varied greatly in size and shape. Therefore, while working
with these GIS-created subregions, a new approach to defining the CD boundaries was
employed. The CDs were overlayed with the PA boundaries and manipulated manually to follow
and not cross PA boundaries, and to follow county boundaries within a PA. This resulted in
each of the 27 PAs being divided into from two to six CDs producing a total of 80 CDs in the
study region. Thus, each CD was comprised of counties for which soil properties information
was aggregated to obtain soil data at the CD level.
To create a single climate data set for each CD, data from multiple climate stations (CS)
were averaged. Initially, all of the CSs that fell within the boundary of a CD were allocated to
the CD for data averaging. However, if there were not enough CSs in a CD, or if their
locations were not geographically representative of the CD, one or more additional CSs were
Figure 7 Climate Station Locations
5-40
-------
Figure 8 Average Maximum Temperature Contours [°C].
"igure 9 Average Annual Precipitation Contours [cm].
5-41
-------
Figure 10 Overlay of Temperature Contours (••••), Precipitation Contours
¦) and Climate Divisions (CD).
Figure 11 Climate Divisions (CD).
5-42
-------
manually allocated. For each CD, the average of the monthly climate values from each of the
allocated CSs were saved in a single file for input to the models.
Agricultural Management.
The RAMS model produces, for each production area, crop rotations and the areas they
occupy. Initially, 80 different crop rotations were identified in the region. Since we are
currently able to model only com, soybeans, wheat, hay (legume and non-legume), millet, and
grasslands, and since we decided it was appropriate to group similar rotations and rotations that
occupied less than 5% of the area in a PA, we were able to reduce the number of crop rotations
to 35. This reduction was also required in order to reduce the number of model runs. Four
tillage practices are considered — conventional till fall plow, conventional till spring plow,
reduced till, and no till. Irrigations and fertilizer applications will be added by the models as
required by the crop physiology depending on the policy options being modeled. Further details
of the modeling methodology and preliminary results of this initial assessment are discussed by
Donigian and coworkers (1993).
SUMMARY
For this work, climate divisions (CD) have been established as the most basic unit to be
modeled (Figure 11). For each CD, soil textures and crop/rotation/tillage (C/R/T) scenarios
have been identified. The next step is to determine the initial conditions of soil carbon and
5-43
-------
nitrogen levels. Initial conditions will be established by making assumptions about the time
extent of various C/R/T scenarios (from » 1900 to 1988), and performing model runs under
those scenarios for a sufficient time while matching historical crop yields through adjustments
in crop, fertilizer, and tillage parameters. The results of the initial condition runs will consist
of soil C and N values for each soil texture in the CD. The soil C values for each texture can
then be weighted by their area distribution within the CD, and summed across all textures to
obtain total C and N values. The results will then be compared with existing soil C and N data.
With initial conditions established, CARD specifies the existing spatial extent of C/R/T for
each production area (PA). These values will be allocated to climate division (CD) C/R/Ts by
multiplying them by the ratio of that CD's area to that PA's area. The base-case model results
will represent a simulation time period of from 1989 to 1990. The model will then be used to
predict soil C and N fluxes and pools to the year 2030, under a variety of policy alternatives
which will include the status quo. The scenarios under consideration include: an increase in the
area of Conservation Reserve Program (CRP) land, an increase in the use of cover crops, and
an increase in the use of conservation tillage. A different policy scenario may generate new
C/R/T practices from RAMS, which will then be used in CENTURY and DNDC to make model
runs for predictions of soil C losses as C02, soil C storages, and total nitrogen losses. DNDC
can also reliably distinguish N20 emissions. Results from the two models can then be compared
and assessments made about the effects of the policy on agricultural management, and about
agricultural management's subsequent effects on soil C and N fluxes and pools [see Donigian
et. al, 1993].
5-44
-------
REFERENCES 1
Barnwell, T. O. Jr., R. B. Jackson, IV, E. T. Elliott, E. A. Paul, K. Paustian, A. S. Donigian,
A. S. Patwardhan, A. Rowell, and K. Weinrich. 1991. Assessment of Methods, Models, and
Databases for Soil Carbon Sequestration Potential for U. S. Agroecosystems. (Unpublished
Report). United States Environmental Protection Agency, Athens, Ga., September 1991.
Barnwell, T. 0. Jr., R. B. Jackson, IV, E. T. Elliott, E. A. Paul, K. Paustian, A. S. Donigian,
A. S. Patwardhan, A. Rowell, and K. Weinrich. 1992. An Approach To Assessment of
Management Impacts on Agricultural Soil Carbon. Water, Air, and Soil Pollution 64:423-435,
August 1992.
Blevins, R.L, G.W. Thomas, M.S. Smith, W.W. Frye and P.L. Cornelius. 1983. Changes in
soil properties after 10 years continuous non-tilled and conventionally tilled corn. Soil & Tillage
Res. 3:135-146.
Campbell, C.A. 1978. Soil organic carbon, nitrogen and fertility. In: Schnitzer, M. and S.U.
Khan (eds) Soil Organic Matter. Developments in Soil Science 8, pp. 173-271. Elsevier Scien.
Pub. Co., Amsterdam.
Center for Agricultural and Rural Development (CARD). 1992. Comprehensive Environmental
and Economic Policy Evaluation System - RAMS Model. Iowa State University. Ames, Iowa
50011.
Donigian, A.S., A.S. Patwardhan, R.B. Jackson, T.O. Barnwell, K.B. Weinrich. 1993.
Modeling the Impacts of Agricultural Management Practices on Soil Carbon in the Central
United States. Advances in Soil Science. (In Press).
Dick, W.A. 1983. Organic carbon, nitrogen, and phosphorus concentrations and pH in soil
profiles as affected by tillage intensity. Soil Sci. Soc. Am. J. 47:102-107.
Goebel, J.J. 1992. National Resources Inventory. USDA Soil Conservation Service. P.O. Box
2890. Washington, D.C. 20013.
Goebel, J.J. and R.K. Dorsch. 1982. National Resources Inventory: A Guide for Users of
1982 NRI Data Files. U.S. Soil Conservation Service. GPO Washington, D.C.
Goebel, J.J. 1987. National Resources Inventory. U.S. Soil Conservation Service. Resource
Inventory Division. USDA Soil Conservation Service. P.O. Box 2890. Washington, D.C.
20013.
1 See references by Barnwell et al., Jackson et al., and Donigian et al. or contact the authors at
their respective addresses for methodology details or to obtain detailed research results.
5-45
-------
Haas, H.J., C.E. Evans and E.R. Miles. 1957. Nitrogen and carbon changes in Great Plains soils as
influenced by cropping and soil treatments. USDA Tech. Bull. 1164. United States Department of
Agriculture, Washington, DC.
Houghton, R., J. Hobbie, J. Melillo, B. Moore, B.J. Peterson, G.R. Shaver, and G.M. Woodwell.
1983. Changes in the Carbon Content of Terrestrial Biota and Soils Between 1860 and 1980: A Net
Release of C02 To The Atmosphere. 53(3) pp. 235-262.
Imhoff, J.C., R.F. Carsel, J.L. Kittle Jr., and P.R. Hummel. 1990. Data Base Analyzer and Parameter
Estimator (DBAPE) Interactive Program User's Manual. EPA/600/3-89/083. Environmental Research
Laboratory, U.S. Environmental Protection Agency, Athens, GA.
Jackson, R.B. 1992. On estimating agricultures net contribution to atmospheric carbon. Water, Air, and
Soil Pollution 64:121-137, August.
Jackson, R.B., A.L. Rowell, and K.B. Weinrich. 1993. Spatial Modeling Using Partially Spatial Data.
Advances in Soil Science. (In Press).
Jenkinson, D.S. and J.H. Rayner. 1977. The Turnover of Soil Organic Matter in some of the Rothamsted
Classical Experiments. Soil Sci. 123:298-305.
Jenkinson, D.S., P.B.S. Hart, J.H. Rayner and L.C. Parry. 1987. Modelling the turnover of organic
matter in long-term experiments at Rothamsted. In: Cooley, J.H. (ed.) Soil Organic Matter Dynamics
and Soil Productivity. INTECOL Bull. 1987:15.
Lamb, J.A., G.A. Peterson, and C.R. Fenster. 1985. Wheat fallow tillage systems' effect on newly
cultivated grassland soils' nitrogen budget. Soil Sci. Soc. Am. J. 49:352-356.
Li, C.S., S.E. Frolking, and T. Frolking. 1992a. DNDC, A model of nitrous oxide evolution from soil
driven by rainfall events: I. Model structure and sensitivity. Journal of Geophysical Research.
97(D9):9759-9776.
Li, C.S., S.E. Frolking, and T. Frolking. 1992b. DNDC, A Model of Nitrous Oxide Evolution from Soil
Driven by Rainfall Events: n. Model Structure and Sensitivity. Journal of Geophysical Research,
97(D9):9777-9783.
Li, C.S., S.E. Frolking, R.C. Harriss, and R.E. Terry. 1992. DNDC Model Application: A Case Study
on N20 Emissions at Belle Grade, Florida. J. of Geophysical Research, (in press).
Partoo, W.J., J. Persson and D.W. Anderson. 1983. Simulation of soil organic matter changes in
Swedish soils. In: W.K. Lauenroth, G.V. Skogerboe and M. Flug (eds.) Analysis of Ecological Systems:
State-of-the-Art in Ecological Systems. Elsevier, New York.
5-46
-------
Parton, W.J., D.S. Schimel, C.V. Cole, and D.S. Ojima. 1987. Analysis of factors controlling
soil organic matter levels in Great Plains Grasslands. Soil Sci. Soc. Am. J. 51:1173-1179.
Parton, W.J., C.V. Cole, J.W.B. Stewart, D.S. Ojima and D.S. Schimel. 1989. Simulating
Regional Patterns of Soil C, N, and P Dynamics in the U.S. Central Grasslands Region. In: M.
Clarholm and L. Bergstrom (eds.). Ecology of Arable Land. Kluwer Academic Press. 99-108.
Parton, W.J., J.W.B. Stewart and C.V. Cole. 1988. Dynamics of C, N, P and S in Grassland
Soils: A Model. Biogeochemistry. 5:109-131.
Paul, E.A., and E.T. Elliot (eds). 1993. Soil Organic Matter in Temperate North American
Agroecosystems. Lewis Press, Ann Arbor, MI. (In Press).
Soil Conservation Service (SCS). 1983. National Soils Handbook. Table 603-45 p603.198
U.S. Department of Agriculture, Soil Conservation Service. GPO Washington, D.C.
Van Der Linden, A.M.A., J.A. van Veen, and M.J. Frissel. 1987. Modelling Soil Organic
Matter Levels after Long-Term Applications of Crop Residues, and Farmyard and Green
Manures. Plant and Soil. 101:21-28.
Vitosh, M.L., J.F. Davis and B.D. Knezek. 1973. Long-term effects of manure, fertilizer, and
plow depth on chemical properties of soils and nutrient movement in a monoculture com system.
J. Envir. Qual. 2:296-299.
Wallis, James R., Dennis P. Lettenmaier, and Eric F. Wood. 1991. A daily hydroclimatological
data set for the continental United States. Water Resources Research. 27(7): 1657-1663.
Wilson, A.T. 1978. The explosion of pioneer agriculture: contribution to the global C02
increase. Nature 273:40-42.
Wood, C.W., D.G. Westfall, and G.A. Peterson. 1991. Soil carbon and nitrogen changes on
initiation of no-till cropping systems. Soil Sci. Soc. Am. J. 55(2):470-476.
ACKNOWLEDGEMENT
This research was performed in part through U.S. Environmental Protection Agency contract
number 68-C0-0019 to AQUA TERRA Consultants, contract number 68-W0-0043 to Computer
Sciences Corporation, and by cooperative agreement CR818652-01-0 to Colorado State
University. Mention of trade names or commercial products does not constitute endorsement
or recommendation for use.
5-47
-------
ASSESSMENT OF THE BIOGENIC CARBON BUDGET OF THE FORMER SOVTET UNION
Tatyana P. Kolchugina, Visiting Research Associate
Ted S. Vinson, Professor
Department of Civil Engineering
Oregon State University
CorvalHs, Oregon 97331
ABSTRACT
A framework was created to quantify the naturai terrestrial carbon eyeie of the former Soviet Union
(FSU). The organization of the caiton cycle parameter and georeferenced data base which support the
framework and the calculations which are required to establish the carbon budget are performed with personal
computer hardware and commercially available spreadsheet software. Based on the framework, net primary
productivity (NPP) for the FSU was estimated at 6.2 ± 1.7 GT (109 tons) C/yr, the vegetation carbon pool
at 118.1 ± 28.5 GT C, the litter carbon pool at 18.9 ± 4.4 GT C, and total soil carbon pool at 404.0 ±
38.0 GT C. The components of the carbon budget obtained with the framework were in good agreement with
estimates from other published sources. The framework will allow the role of the FSU in the global carbon
cycle to be assessed. The extent of forest and agricultural ecosystems within the FSU that can be technically
managed on a sustainable basis to conserve and sequester carbon may also be determined with the framework.
Although the research described in this article has been funded under U.S. Environmental Protection
Agency Agreement CR17682-01 to Oregon State University, it has not been subjected to the Agency's review
and, therefore, does not necessarily reflect the views of die Agency, and no official endorsement should be
inferred.
INTRODUCTION
The long-term ecological consequences of the change in the chemical composition of the atmosphere
are not fully understood; however, a warmer global climate is highly probable (1). If C02 concentrations
were to double, the earth's temperatures may rise between 1 and 5*C (2). Climatic changes may be more
pronounced in the Northern Hemisphere (3). Global warming may accelerate the rates of plant respiration
(4) and decay of organic matter (5).
In view of the potential to significantly disrupt the equilibrium of the natural carbon cycle, it may be
necessary to offset increased amounts of atmospheric COi. This will require the development of international
strategies to reduce.industrial emissions or manage terrestrial carbon stores. Before any international strategy
can be formulated, policies aimed at maintaining a desirable carbon balance within national boundaries would
be required. The determination of a carbon balance includes the quantification of the natural (biogenic) and
5-48
-------
anthropogenic contributions to the carbon cycle within national boundaries. The quantification of the carbon
cycle following an assessment of carbon pools and fluxes is generally referred to as the carbon budget.
Carbon budgets recently have been established for Sweden (6) and the forest sectors of Canada (7) and the
United States (8).
The former Soviet Union (FSU) was the largest country in the world. It occupied one-sixth of the land
surface of the earth. An understanding of the carbon budget of the FSU is essential to the development of
international strategies aimed at mitigation of the negative impacts of global climate change.
Anthropogenic carbon emissions in the FSU have recently been estimated at approximately 1 GT (9).
Despite an abundance of Soviet data on carbon-cycle parameters in desert, tundra, forest and grassland
ecosystems (10-18 and others), the national carbon budget of the FSU, specifically the natural component,
has only recently been established, as described herein.
THE PHYSICAL ENVIRONMENT OF THE FSU
The territory of the FSU is represented by a variety of climate conditions. The major pan of the
territory is in the temperate climatic zone. The climate in the FSU changes from arctic and subarctic in the
North to subtropical and desert in the South. From west to east, the climate makes a transition from maritime
to continental to monsoon.
Matthews (19) identified eight principal types of vegetation in her global data base: forest, woodland,
shrubland, grassland, tundra, desert, marsh/swamp and cultivated land. The vegetation of the FSU is
represented by a variety of formations, including all of these major types. Arctic deserts and tundra
formations are found in the northern regions of the FSU; deserts and semi-deserts occur in southern regions.
A vast area, the largest of any country in the world, is occupied by forests and grasslands. Hie total area
of forest zone under State supervision (in 1983) is 1,259 million ha (20), which is approximately 56.5 percent
of the territory of the country. About 95 percent of the forest area is in Russia. Tundra and boreal forests
store a significant amount of organic matter. Twenty-seven percent of the approximately 80 percent of
terrestrial organic matter stored in soil is found in boreal ecosystems (21). Grasslands are also an important
component of die terrestrial carbon cycle. Despite the fact that grasslands do not accumulate large quantities
of plant mass (compared with forest ecosystems), they exhibit high net primary productivity (NPP) and,
therefore, may influence the terrestrial carbon cycle.
Peat lands, which are wetlands where peat is accumulating, store a significant amount of carbon.
Organic soil carbon content reaches 2,000 t/ha (22). The FSU has the greatest expanse of peat lands in the
world (23). Wetlands are known to be a source of methane to the atmosphere (24-26). Although the
atmospheric concentration of CH4 is much town- than the concentration of CO2, CH* is 20 times more
effective (per molecule) than CO2 as a greenhouse gas (27).
THE NATURAL (BIOGENIC) TERRESTRIAL CARBON CYCLE
Hie natural (biogenic) terrestrial carbon cycle consists of a combination of carbon pools and fluxes as
shown in Figure 1. The pools are carbon stores in soil and vegetation, including living vegetation (i.e.,
phytomass) and plant detritus (i.e., mortmass and litter). In the present study, the term mortmass was used
to describe coarse above-ground and below-ground woody debris. The term litter was used to define the
upper soil layer comprised of fine woody debris and leaves that are not completely decomposed. The effluxes
5-49
-------
&L
rii i'
Equilibrium
Assumption:
F = F - F
4 3 12 SM
Ft-F^-F.-F,
PS
f,«f7
Pools
Pp„
- Phytomass
P*
- Mortmass
P,
- Litter
P,
-SoU
PP5
- Proteaed soil
organic matter
IbQmzs
Fj - Net Primary
Productivity (NPP)
F2 - Mortmass formation
(stem, branch, & root
detritus)
F4 - Foliage formation
(Le. leaf litter fall)
m±
- Total formation of
soil organic matter
(from mortmass [M]
and litterfL})
- Formation of protected
organic matter
to
Effluxes
- Mortmass
decomposition
- Liner
decomposition
- Unprotected Soil
organic matter
decomposition
- Protected soil
organic matter
decomposition
- Autotrophic
respiration
(F, + F,, = GPP)
Figure 1. Natural (biogenic) terrestrial carbon cycle.
5-50
-------
are carbon emissions resulting from plant respiration and decomposition of organic matter. The processes
of formation of new organic matter in soil and vegetation (i.e., humus and foliage formation) and NPP
represent carbon influxes.
The NPP equals the difference between gross photosynthesis (GPP) and respiration of autotrophic
organisms (RA). The RA amounts to 44 to 52 percent (48 percent on average) (28) of GPP. Root respiration
(RAr) comprises one-third of the RA. An ecosystem sequesters carbon during a specific phase of development
(29). The parameter that characterizes this carbon storage is called net ecosystem productivity (NEP). The
NEP equals the difference between NPP and the carbon loss resulting from heterotrophic respiration (Ry).
Carbon fluxes can be measured or calculated. However, when carbon effluxes are measured, the
contribution from different processes cannot be distinguished. For example, when soil carbon efflux is
measured, it is difficult to distinguish between effluxes resulting from RAr and RH (decomposition of litter,
below-ground mortmass and soil organic matter). The quantitative method allows one to separate fluxes.
METHODOLOGY TO ESTABLISH THE NATURAL (BIOGENIC) CARBON BUDGET
To establish the natural (biogenic) carbon budget, the geographic area within which it was quantified
was isolated. The term ecoregion was applied to the boundaries and areal extent of the geographic area. The
term ecosystem was applied to the combination of certain soil-vegetation formations within an ecoregion.
The concept of an ecosystem is a broad one; its function is to emphasize obligatory relationships and
interdependence (29). The term biome was applied to the complex of ecosystems within a climatic belt or
subbelt. Nine biomes within the FSU were identified: polar deserts, tundra, forest-tundra/sparse taiga, taiga,
mixed-deciduous forests, forest-steppe, steppe, desert-semi-desert and subtropical woodlands.
Two approaches may be used to identify ecoregions, namely, the use of: 1) maps with specific
information, and 2) satellite imagery and remote-sensing techniques. A combination of these approaches may
also be used. For example, ecoregions identified by map work may be validated with satellite imagery and
remote-sensing techniques. At the present time (1991), several maps were used to isolate the ecoregions
within which the natural carbon cycle was quantified. In the continuation of the research program, satellite
imagery and remote-sensing methods will be used to validate the ecoregions defined for the FSU.
Carbon cycle parameters have been quantified by soil, agricultural and forest scientists, ecologists and
botanists for several decades. The carbon cycle parameters may be expressed in terms of carbon content (for
pools) or rate (for influxes or effluxes)/ha for a variety of soil-vegetation complexes. If die soil-vegetation
complexes are related to the natural attributes identified on maps, which are used to isolate ecoregions, the
carbon budget for an ecoregion can be established simply by multiplying the area of the ecoregion (in ha) by
the carbon content(s) and flux(es). The carbon contents and fluxes for all the ecoregions may be summed
to arrive at the carbon budget for a larger region, biome or nation.
Based on the preceding discussion, the framework shown in Figure 2 was created to assess a natural
component of the carbon budget. Initially, maps were used to isolate ecoregions (Frames I and 2) and data
bases, which contain natural carbon cycle parameters, were compiled (Frame 3). The areal coverage of the
ecoregions was integrated with the carbon content and flux data bases to establish the carbon budget within
the ecoregion (Frame 4). The organization of the carbon cycle parameter data base, the ha data for the
ecoregions and the calculations that were required to establish the carbon budget were performed with
personal computer hardware and commercially available spreadsheet software (30) in a Windows"
environment. The carbon budgets for the ecoregions were summed to establish the carbon budget for a biome
5-51
-------
Computer Superimposition
to Identify Ecoregfons within
Georegions
Data Bases for
Natural Carbon Cycle
(Bazilevich, 1986)
(Kobak, 1988)
Integration of Ecoregion data to
assess Carbon Pools and Fluxes for
(Georegion Spreadsheet Format)
Integration of Ecoregion data to
assess Carbon Pools and Fluxes for
Biomes within the Soviet Union
(Summary Spreadsheet Format)
Collect & Digitize Maps
(Ryabchikov,1988)
(Isachenko.1988)
(Cherdantsev, 1961)
(Kolchugina, 1991;
after Gerasimov, 1933,
and Bazilevich, 1986)
Estimate of Carbon Pools and
Fluxes for Ecoregions
(Georegion Spreadsheet Format)
Figure 2. Framework to assess the natural (biogenic) component of the carbon budget
5-52
-------
or the entire territory of the FSU (Frames 5 and 6). The specific activities related to the execution of these
steps are discussed in the following paragraphs.
ISOLATION OF ECOREGIONS (FRAMES 1 AND 2)
About 95 percent of the territory of the FSU, including Russia, Ukraine, Belonissia, Kazakhstan and
the Baltic states, was categorized by the soil-vegetation type of the ecosystem, the presence of wetlands and
cultivation intensity. Maps containing information on the distribution of zonal soil-vegetation associations
within the FSU (31), distribution of wetlands (32) and cultivation intensity of arable lands (33) were digitized
and computer-superimposed with a geographical information system (GIS) (34). The map with the
distribution of soil-vegetation associations (31) provided the basis for ecoregion isolation. In addition, eight
georegions (Near Ocean; Eastern, Middle and Western Siberia; Eastern, Central and Western Europe; and
Kazakhstan) were defined to accommodate geodependence of carbon accumulation (17,35). These georegions
were also mapped, digitized and computer-superimposed with the GIS. After computer superimposition of
the four maps noted above, more than 70 ecoregions related to different ecosystems (i.e., soil-vegetation
associations) were identified. Wetland, floodplain, mountain ecosystems and arable land were isolated within
these ecoregions. The ecosystems presented by Ryabchikov (31) were aggregated into nine biomes.
The map used to isolate wetlands (32) allows one to determine the total area of wetlands, but does not
allow different wetland landscapes to be distinguished. There are at least three main classifications of mire
systems in the FSU (36). According to the trophic conditions and the developmental stage, mires can be
classified as eutrophic, mesotrophic or oligotropbic. According to the hydrological conditions, mires may
be divided into minerotrophic (both ground and rain water supply) and ombrotrophic (rain water supply).
According to the main layer of plant communities, mires can be divided into moss, graminoid, dwarf-shrub
and shrub. Carbon accumulation differs depending on wetland type. Further studies will require the
incorporation of maps that would allow one to isolate different types of wetlands. However, for the present,
the superimposition of wetland and soil-vegetation maps allowed the determination of wetland type within a
given ecosystem and the specification of carbon accumulation parameters.
DATA BASES FOR NATURAL CARBON CYCLE PARAMETERS (FRAME 3)
Bazilevich (17) compiled a data base on carbon accumulation in vegetation from studies of 1,500
vegetation complexes in the FSU. The data base is a comprehensive source of information on all vegetation
formations in the FSU, namely, 13 polar desert and tundra, 40 forest, 57 grassland, 20 mire ecosystems and
more than 50 desen-semidesert formations. These vegetation complexes were correlated to the ecosystems
presented by Ryabchikov (31). The data base provides site-specific values for total phytomass content and
phytomass productivity for all vegetation formations in the FSU. The data base allows the assessment of
phytomass and phytomass increment allocation. Phytomass was categorized as green-assimilating, woody
above- (stems and branches) and below-ground (roots and buried stems) parts of plants. Moraoass was
categorized as woody above- (dead stems, branches, grass, windfall), and below-ground parts of plants and
litter. Productivity of phytomass or NPP was categorized in the same manner as phytomass. The net carbon
content of plant mass was assumed to be 50 percent (28). This percentage was used to calculate the net
carbon storage and rates of carbon accumulation in vegetation.
Kobak's (28) data base was used to characterize the soil component of the carbon cycle. The data base
resulted from the analysis of the published soil data. About 70 different Soviet and foreign sources were
included (22,37-40, and others). Soil carbon cycle parameters for more than 40 soil types of polar, boreal
and tropical belts represent the averages from the empirical data presented in different sources. However,
in some cases, only certain data are taken into account. For example, the Soviet data are used to characterize
5-53
-------
the carbon contents of podzol and chernozem soils. The data base includes carbon contents of soil (total and
stable portion; in one-meter layer for inorganic soils and in all thickness for the bog organic soils), the annual
rate of foliage and humus formation (total and stable portion), and CO2 efflux from soils.
Matthews and Fung (41) compiled a data base on methane emissions from natural wetlands. Three
sources: 1) vegetation classified according to Matthews (19); 2) soil properties (42); and 3) fractional
inundation from a global map survey of Operational Navigation Charts (ONC) are used to assess the global
distribution of wetlands. Published data are analyzed to compile 1) typical methane emissions for wetland
ecosystem type and 2) length of the active season (24,25,43,44, and others). Data on methane emissions
from wetlands (41) were correlated to the data on wetland distribution within the nine biomes in the FSU.
CORRELATION OF DATA BASES TO MAPPED ECOSYSTEMS (FRAME 4)
The maps and data bases are not specifically designed for carbon cycle quantification. However, the
names of soil-vegetation associations reported by Ryabchikov (31) corresponded well with the descriptions
of vegetation formations given by Bazilevich (17) and soil types given by Kobak (28).
INTEGRATION OF ECOREGION AREAS AND CARBON DATA BASES (FRAME 5 AND 6)
Carbon pools and fluxes for natural ecosystems in the FSU were estimated by integrating the carbon
data bases and the GIS analysis results (hectare data) within commercially available spreadsheet software (30).
Productivity of green-assimilating parts of plants was used to characterize the rate of foliage formation. Data
on carbon content in bog soils were integrated with ha data on the extent of wetlands within biomes. Low,
mean and high estimates were made for each of the eight georegions by summing the contributions from each
ecoregion. The carbon pools and influxes for the forest biomes in the FSU were obtained by summing the
georegion totals for the nine biomes. Initially it was assumed that 1) natural ecosystems are presently in a
state of equilibrium, (NEP) equals 0; 2) forests totally cover the area of ecosystems (excluding arable land)
within forest-tundra/sparse taiga, taiga and mixed-deciduous forest biomes; and 3) forests occupy one-half
of the area (excluding arable land) of the forest-steppe biorae.
Carbon effluxes were calculated from the influxes assuming that all ecosystems were initially in an
equilibrium state (NPP equal to RH). Mortmass decomposition was assumed to be equal to mortmass
production. In turn, mortmass production was assumed to be equal to phytomass production (NPP and
production of different parts of plants). Carbon efflux from litter decomposition was calculated as the differ-
ence between foliage formation (green-assimilating parts production) and the sum of total humus formation
and peat accumulation. Carbon effluxes from soil organic matter decomposition were calculated as the
difference between total and stable humus formation. Carbon efflux from *Ar wis calculated from the NPP,
assuming that RAr comprises one-third of the total RA> and RA comprises 48 percent (on average) of die
GPP; NPP equals the difference between GPP and RA. The sum of R^ and RH (below-ground mortmass,
litter, and soil organic matter decomposition) may be compared with field measurements of the surface soil
carbon efflux (28).
The estimates of carbon cycle parameters obtained in the present study were validated, where possible,
with the help of other data bases. The global data base for the NPP of terrestrial ecosystems, compiled with
the help of advanced very high resolution radiometer (AVHRR) data (45), was incorporated in the study.
The NPP for natural (non-arable) ecosystems was calculated by dividing NPP totals for the ecoregion by the
total number of ha and multiplying by the number of ha of natural ecosystems within the ecoregion.
5-54
-------
CARBON POOLS AND FLUXES IN THE FSU ... EXAMPLE RESULTS
The vegetation carbon pool of natural ecosystems in the FSU was estimated at 118.1 ± 28 J GT. The
vegetation carbon pool included phytomass (91.0 ± 22.0 GT) and raortmass (27.1 ± 6.5 GT). The litter
carbon pool was estimated at 18.9 ± 4.4 GT. The soil carbon pool was estimated at 404.0 ± 38.0 GT
(including peat), with 269.0 ± 25.0 in the stable form. Peadands accumulated 148.0 GT C.
The totaJ productivity of phytomass of the nine biomes was estimated at 6.2 ± 1.7 GT C/yr. The
productivity of green assimilating parts based on Bazilevich's (17) data was estimated at 2.5 ± 0.6 GT C/yr.
Humus formation was estimated at 257.0 ± 94.0 MT (10^ t) C/yr (87.8 ± 15.1 MT C/yr in stable form).
Methane production in peat lands was estimated at 7.6 ± 1.6 MT C/yr. FSU biomes may be arranged in
four groups based on the distribution of biomass (phytomass and mortmass): 1) polar-desert biome; 2) tundra
and forest-tundra/sparse taiga biomes; 3) taiga, mixed-deciduous forest, forest-steppe and subtropical
woodlands biomes; and 4) steppe and desen-semidesert biomes.
In the polar desert biome, mortmass was greater than phytomass and was distributed equally above and
below ground. Above-ground phytomass was significantly greater than below-ground phytomass. In the
tundra and the forest-tuncrs/sparse-taiga biomes, mortmass also exceeded phytomass. In the tundra biome,
below-ground biomass was more developed. In the forest-tundra/sparse-taiga biome, phytomass and
mortmass were distributed in equal proportions above and below ground.
In the forest biomes, phytomass was greater than mortmass. Above-ground pans were greater than
below ground parts in both pools. Below-ground phytomass was approximately the same as above-ground
mortmass. In the steppe and desert-semidesert biomes, phytomass and mortmass did not differ substantially.
Below-ground pans were greater than above-ground parts in both pools. In the steppe biome, above-ground
mortmass was greater than above-ground phytomass, while in the desen-semidesert biome, more living pans
of plants were found above-ground.
The NPP of natural ecosystems estimated with the data base from Fung et al. (45) was compared with
the estimate of phytomass productivity based on data given by Baziievich (17). Total phytomass productivity
(6.2 GT C/yr) estimated from Baziievich (17) was about 2.2 times higher than NPP (3.0 GT C/yr) obtained
from Fung et al. (45) (the polar desen biome represented one exception: NPP obtained from Fung et al. (45)
was greater than phytomass productivity obtained from Baziievich (17)). The NPP obtained from Fung et
al. (45) corresponded well only with the productivity of green-assimilating parts given by Baziievich (17).
The discrepancy may be due to the fact that either root productivity or both root and above-ground woody
phytomass productivity are underestimated by Fung et al. (45). Further, a discrepancy in the results could
be related to the use of different methods to identify carbon-quantifiable regions. The rate of foliage
formation estimated from Kobak's (28) data was 3.6 ± 0.6 GT C/yr. The value was slightly greater than
estimates based on data from Baziievich (17).
The carbon efflux from the litter decomposition estimated from the equilibrium analysis was 2.3 GT/yr.
Soil organic matter decomposition was 0.2 GT C/yr. The was estimated at 1.8 GT C/yr. Below-ground
mortmass decomposition was estimated at 2.5 GT C/yr.
The carbon efflux resulting from litter and soil organic matter decomposition (2.4 GT C/yr) was less
than the soil surface CO7 efflux (3.6 ± 1.1 GT C/yr) estimated for the nine biomes based on data reported
by Kobak (28). The aggregated efflux from *Ar and average efflux resulting from the decomposition of
5-55
-------
below-ground mortmass, liner, and soil organic matter was estimated as 6.7 GT C/yr. This figure was 30
percent greater than the maximum COj efflux (4.7 GT C/yr) estimated from Kobak's (28) data.
Based on Avogadro's number (6.028 x IQ23 molecules in a mole), methane emissions from peat lands
were equivalent to 3.8 x 1035 molecules of CH4. The total efflux from below-ground mortmass, litter and
soil organic matter decomposition was equivalent to 3,437 x 1033 molecules of COs- Assuming CH4 is 20-
fold more active than CO2 as a green-house gas, methane emissions represented approximately 2.2 percent
of the soil surface efflux estimated in the present study.
AGRICULTURAL MANAGEMENT OPTIONS TO CONSERVE AND SEQUESTER CARBON
The cultivated area of the FSU was 211.5 million hectares in 1988 (46). This huge area is nearly twice
the cultivated area of the United States. Clearly, the agricultural sector of the FSU may play a significant
role in the global carbon cycle. "Hie following technical management options to conserve and sequester
carbon in the agricultural sector of the FSU have been identified by Gaston, et ai. (47) and Rozhkov (48):
* Conservation tillage slows the loss of soil carbon by reducing erosion and by decreasing
soil temperature and increasing soil moisture, thus reducing the oxidation of carbon. In
substantial areas of the FSU, coid temperatures and high soil moisture may limit the use
of this option.
* The addition of organic wastes has been shown to increase soil organic carbon (49). The
potential to realize this option is relaxed to the geographic distribution of livestock and
current efficiency in returning manure to the soil.
• Cover cropping (or green manuring) improves soil structure and increases organic carbon
content (50). Data which would allow the increase in carbon content by soil type and crop
type is not available.
• The benefits of agrofbrestry (i.e.. Integrating trees with agricultural lands) are widely
recognized. In the dry, windy conditions of the southern FSU the planting of tree belts
(shelter belts) or hedges would enhance agricultural production, reduced soil erosion and,
therefore, result in the direct sequestration of carbon.
FORESTRY MANAGEMENT OPTIONS TO CONSERVE AND SEQUESTER CARBON
A. Shwidenko (SI) and Krankina, et. al. (52) have identified several management options that may
increase the carbon storage or reduce the carbon source of forest ecosystems in the FSU, as follows:
* Reforestation of 95 million hectares of unforested area of the State Forest Fund (from 145
million hectares theoretically available for reforestation)
Increased fire control and prevention would substantially reduce the 30 thousand forest fires
which occur annually within the FSU. If the extent of forest fires could be decreased by
50 percent, a substantial amount of carbon would be sequestered.
5-56
-------
• Enhanced foresr nroductivitv could be achieved mainly by reconstruction of low-stocked
stands, partial replacement of deciduous stands with long-iiving coniferous, and
replacement of climax stands.
• Enhanced disease and pest control includes the decrease and subsequent elimination of pests
and diseased areas that presently comprise three million hectares, removal and selective
cutting of dead and sick trees, etc.
• Reconstruction of the forest-industrial sector includes mainly a transition to ecologically-
protective methods of wood harvesting that would allow rapid regeneration of forest stands,
and an increase in efficiency of wood utilization from 50 (present assessment) to 70
percent.
• Increase in stand harvest age of main cuttings in the European part of the FSU is
potentially interesting, theoretically realistic, and would result in substantial additional
sequestration of carbon. However, high wood demand in the FSU and other countries
might prevent the implementation of this option.
These options will depend on the forest policy of the republics of the FSU. To assess the additional amount
of carbon that may be sequestered by implementation of these management options the trends in national
forest policy should be known together with forest statistical data which reflects existing management
practices for Soviet forests. Forest statistical data are available in the Forest Fund of the USSR (S3).
SUMMARY AND CONCLUSIONS
A framework was created to quantify carbon pools and fluxes in the FSU at the ecosystem, biome,
georegional or national scales. Hie organization of the carbon cycle parameter and geo referenced data base
which support the framework and the calculations which are required to establish the carbon budget are
performed with personal computer hardware and commercially available spreadsheet software. The
components of the biogenic carbon budget obtained with the framework were in good agreement with
estimates from other published sources. The greatest advantage of the framework is that all elements
identified may be easily updated and the carbon budget can be recalculated immediately thereafter. Sensitivity
analyses may also be performed with the model. The framework will allow the role of the FSU is the global
carbon cycle to be assessed. The extent of forest and agricultural ecosystems within the FSU that can be
technically managed on a sustainable basis to conserve and sequester carbon may also be determined with the
framework.
REFERENCES
1. Intergovernmental Panel on Climate Change (IPCC). Houghton. TJ., Jenkins, G J., and Ephraums,
J J. (eds.), Climate change, the IPCC scientific assessment. Working group 1 report, WMD & UNEP
Univ. Press, Cambridge, UK, 1990. 365 pp.
2. Schneider, S.H. Global warming. Vintage Books, New York, 1990.
3. Etkin, D. Greenhouse warning: consequences for Arctic climate. Tnumal of Cold Regions
Engineering. Amer. Soc. of Civil Engineers, 4(l):54-66, 1990.
5-57
-------
4. Keeiing, C.D., Bacastow, R.B., Carter A.F., Piper S.C., Whorf, T.P., Heimann, M., Mook, W.G.,
and Roeloffeen, H. A three-dimensional model of atmospheric C02 transport based on observed winds:
1. Analysis of observational data. Geographical Monograph 55:165-236, 1919,
5. Dixon, R.K. and Turner, D.P. The global carbon cycle and climate change: responses and feedbacks
from below-ground systems. Environmental Pollution 73: 245-262, 1991.
6. Eriksson, H. Sources and sinks of carbon dioxide in Sweden. AMBIO 20(3-4): 146-150, 1991.
7. Apps, MJ., Kurz. A.W., and Price, T.D. Application of a carbon budget model to strategic planning
for the effects of climate change on the Canadian forest sector, (this volume), 1991.
8. Gucinski, H., Turner, D.P., and Koerper, G. The carbon flux on forested lands of the United States,
in: Proceedings of the IUFIO Conference on Integrating Forest Information over Space and Time,
Canberra, Australia, 1992.
9. Makarov, A.A. and Bashmakov, I. 1990, The Soviet Union: Carbon emission control strategies.
In: William U. Chandler (ed.), Case Studies in International Cooperation, 1990.
10. Baziievich, N.I., Rodin, L.Ye., and Rozov, N.N. Geographical aspects of biological productivity
studies. In: Proceedinp of 5th Congress of Geographical Society of USSR. Leningrad, 1970.
11. Kazimirov, N.I. and Morasova, R.M. Biological turnover of substances in spruce forests of Karelia,
Nayka, Leningrad, 1973.
12. Aleksandrova, V.D. Geobotanical regionalization of Arctic and Antarctic, Nauka, Leningrad, 1977.
13. Tytlyanova, A.A. Biological cycle of carbon in herbaceous formations, Nauka, Novosibirsk, 1977.
14. Vatkovskyi, O.S.The analysis of primary production of form formations, Nayka, Moscow, 1976.
15. Dylis, N.V. and Nosova, L.M. Phytomass of forest biocenosis. Nauka Press, Moscow, 1977.
16. Kazantseva, T.I. Productivity and dynamics of above-ground phytomass of deserts. Problems of desert
util Ization 2:76-33, 1980.
17. Baziievich, N.I. Biological productivity of soil-vegetation formations in the U.S.S.R. Bulletin of
Academy of Sciences nf the (T S S.R. Geographical Series 2:49-66. 1986.
18. Baziievich, N.I., Grebensbikov, O.S., and Tyshkov, A.A. Natural association of geographical patterns
with the structure and functioning of ecosystems, Nayka Press, Moscow, 1986.
19. Matthews, E. Global vegetation and land use: new high-resolution data bases for climate studies.
Journal of Climate and Annlied Meteorology 22:474*487. 1983.
20. Vorobyov, G.I., ed. Forest encyclopedia. Moscow, 1985.
21. Billings, W.D. Carbon balance of Alaskan tundra and taiga ecosystems: past, present and future.
Quaternary Science Review 6:165-177. 1987
5-58
-------
22. Bohn, H.L. Estimates of organic carbon in world soils: II. Journal of Soil Science Society of Arizona
4:1118-1119, 1982.
23. Tyuremnov, S.N. Torfyanye mestorozhdeniya. Nedra, 3rd ed., 1976.
24. Barxlett, K.B., Bartlett, D.S. and Sebacher, D.I. Methane flux from coastal salt marshes. Journal of
Geophysical Research 90:5710-5720, 1985.
25. Bartlett, K.B., Bartlett, D.S., Sebacher, D.I., Harris, R.C., and Brannon, D.P. Sources of
atmospheric methane from wetlands. Paper presented at the 36th Congress of the International
Austronautic Federation, Stockholm, Sweden, October 7-12, 1985.
26. Harriss, R.C., Gotham, E., Sebacher. D.I., Banlett, K.B., and Flebbe, P.A. Methane flux from
northern peatlands. Nature 315:652-653. 1985.
27. Blake, D.R. and Rowland, F.S. Continuing worldwide increase in trapospheric methane, 1978-1987.
Science 239:1129-1131. 1988.
28. Kobak, K.I. Biotical compounds of carbon cycle. Hydrometeoizdat, Leningrad, 1988.
29. Odum, E.P. Fundamentals of Ecology. W.B. Saunders C., Philadelphia-London, 1953.
30. Microsoft Corporation. Microsoft Excel, Redmond, Washington, 1991.
31. Ryabchikov, A.M., ed. Map: Geographical belts and zonal types of landscapes of the world. School
of Geography, Moscow State University, Moscow, 1988.
32. Isachenko, A.G., ed. Landscape map of the U.S.S.R. Institute of Geography, Leningrad State
University, 1988.
33. Cherdantsev, G. N. Map: Arable land in the U.S.S.R. in 1954. In: A Geography of the U.S.S.R.
— Background to a Planned Economy edited by J.P.Cole, F.C. German, 290 pp, 1961.
34. Burrough, P.A. Principles of geographical information systems for land resources assessment.
Clarendon Press, Oxford, New York, 1986. 193 pp.
35. Gerasimov, I.P. About soU-ciimaticai faces of the flatlands of the U.S.S.R. and adjacent countries,
la: Proceedings of the V.V. Docuchayev Soil Institute. Vol. 8, No. 5, U.S.S.R. Academy of Sciences
Press, Leningrad, 1933.
36. Botch, M.S. and Masing, V.V. Mire ecosystems in the U.S.S.R. In: Ecosystems of the World:
Mires, Swamp, Bog, Fen, and Moor, Vol. 4B, Regional Studies, edited by A J.P. Gore, Elsevier, New
York, 1983. pp. 95-152.
37. Kononova, M.M. Soil organic matter, Nauka Moscow, 1963.
38. Schlesinger, W.H. Carbon balance in terrestrial detritus. Ann. Rev. Ecolog. Svst 8:51-81, 1977.
5-59
-------
39. Post, W.M., Emanuel, W.R., Zinke, P.I., and Stagenberger, A.G. Soil carbon pools and world life
zones. Nature 298(5870): 156-159, 1982.
40. Kobak, K.I. and Kondrashova, N.Yu. Distribution of organic carbon in soils of the globe. TrudvGGI
320: 61-76, 1986.
41. Matthews, E. and Fung, I. Methane emission from natural wetlands: global distribution, area, and
environmental characteristics of sources. Global Biogeochemical Cycles i(l):61-86, 1987.
42. Zobler, L. A world soil file for global climate modeling, NASA Technical Memo 87802, 1986.
43. Svensson, B.H. Carbon dioxide and methane fluxes from the ombrotrophic parts of a subarctic mire.
Ecological Bulletin (Stockholm) 30:235-250, 1980.
44. Sebacher, D.I., Harris, R.C., Bartlett, K.B., Sebacher, S.M., and Grice, S.S. Atmospheric methane
sources: Alaskan tundra bogs, an alpine fen, and a subarctic boreal marsh. Tellus 38B: 1-10, 1986.
45. Fung, I.Y., Tucker, C J., and Prentice, K.C. Application of advanced very high resolution radiometer
vegetation index to study atmosphere-biosphere exchange of CO^ Journal of Geophysical Research
92(D3):2999-3015, 1987.
46. USDA. USSR Agriculture and Trade Report Situation and Outlook. USDA RS-90-1, May 1990.
47. Gaston, G.G., Kolchugina, T.P., and Vinson, T.S. Potential effect of conservation tillage on carbon
in agricultural soils of the former Soviet Union. (Submitted to Journal of Agriculture. Ecosystems, and
Environment.)
48. Roshkov, V. Deputy Director, Dokuchaev Soil Institute, Moscow, Russia. (Personal communication)
1992.
49. Poyser. E.A., Hedlin, R.A., and Ridley, A.O. The effect of farm and green manure on the fertility
of black Earth meadow clay soils. Canadian Journal of Soil Science. 37:48-56, 1957.
50. MacRae, R J. and Mehuys, G.R. Hie effect of green manuring on the physical properties of temperate
soils in Advances in Soil Science, Vol. 3. Springer-Verlag, New York, 1985. pp. 71-94.
51. Schwidenko, A.Z. Director of Forest Resources Institute, State Forest Committee, Moscow, Russia.
(Personal communication) 1992.
52. Krankina, O., Dixon, R., Kolchugina, T., and Vinson, T. Forest management options to conserve and
sequesterterrestrial carbon in Russia. (In preparation) 1992.
53. Forest Fund of the USSR. USSR State Forestry Committee, Vol. 1, pp. 1005, 1990.
5-60
-------
HOUSEHOLD FUELS IN DEVELOPING COUNTRIES:
GLOBAL WARMING, HEALTH, AND ENERGY IMPLICATIONS
by
Kirk R. Smith
Program on Environment
East-West Center, Honolulu, HI 96848
and
Susan A. Thomeloe
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
ABSTRACT
Although individually small, the widespread and daily use of household stoves
with poor combustion efficiency in developing countries Taises questions about possible
global warming and other environmental implications of their airborne emissions. To
explore the possible utility of efforts to measure the emissions from representative
samples of these devices, a small pilot study of greenhouse gas emissions of biomass and
fossil-fuel stoves was undertaken in Manila * (Smith et al„ 1992a&b). The results,
although based on only a few measurements, indicate that such stoves may have a
significant role in global greenhouse gas inventories; be subject to substantial
improvement through alternative technologies; and that policy measures should
consider energy and health implications as well. As a consequence, a larger set of
studies is being planned for India, China, Thailand, and Brazil.
This research is funded through EPA's Global Climate Change Research
Program. Research on emissions and mitigation of major sources of greenhouse gases
is being conducted by EPA's Air and Energy Engineering Research Laboratory
(AEERL). This paper has been reviewed in accordance with the EPA's peer and
administrative review policies and approved for presentation and publication.
l Organized by the East-West Center, households were selected as pan of a study of urban energy
and air pollution funded by the International Development Research Center, Ottawa; physical
sampling, laboratory analyses, and data reporting were supported by die U.S. Environmental
Protection Agency; and additional data analysis and evaluation were funded by the Energy Sector
Management Assistance Program, Worid Bank, and the United Nations Development Program.
Collaborators included the College of Engineering of the University of the Philippines, the Oregon
Graduate Institute, and Lawrence Berkeley Laboratory of the University of California. We greatly
appreciate the contributions and suggestions of our colleagues in these institutions (Smith et aL,
1992b).
5-61
-------
INTRODUCTION
It has been said that wood is the fuel that heats you twice, once when you chop
it and again when you bum iL Like fossil fuels, however, biofuels also have the
potential to heat you a third time as a result of enhanced greenhouse wanning due to the
gases released by combustion. It has generally been assumed that this potential is
realized only when the biomass being burned is harvested on a non-sustainable basis.
With sustainable harvesting, it is argued, an equivalent amount of carbon is recaptured
by the re growing biomass as released by combustion. Thus, the net greenhouse gas
increment is zero. Even when this is true with regard to the number of carbon atoms,
however, it may not be with regard to their greenhouse equivalence. In particular,
photosynthesis captures only carbon dioxide (CO2) from the atmosphere, but actual
biomass combustion emits other carbon-containing materials including molecules other
than CO2 with atmospheric wanning impacts (Levine, 1991).
These products of incomplete combustion (PICs) are also of concern because of
their effects on human health. In many parts of the U.S., for example, smoke from
wood-fired heating stoves is the principal cause of some types of ambient pollution
during much of the year. For the nation as a whole, wood combustion is a major
emissions source for some important air pollutants, such as particulates and polycyclic
aromatic hydrocarbons.
Most of the world's woodfuel and other forms of biofuel, such as crop residues
and animal dung, however, are burned not in metal heating stoves in developed
countries, but in simple open cookstoves in developing countries. Approximately half
the households in the world cook in this fashion. Measurements in collage homes
throughout the world have shown that health-impairing concentrations of PICs are
often encountered where people use wood or other biomass for cooking or heating
under poorly ventilated conditions (Smith, 1987).
These same PICs also represent lost energy and contribute to the low engineering
efficiency with which meals are cooked in much of the developing world (Baldwin,
1987). This in turn increases pressure on biomass resources, which, along with land
clearing and other factors, has been associated with deforestation and accompanying
environmental problems in some areas.
The apparent opportunity for decreasing forest-stressing biofuel demand as well
as reducing health-threatening smoke exposures has lured many local, national, and
international organizations, both government and private, into programs to disseminate
improved biomass stoves in poor countries. Although there have been major successes,
such as the Chinese national improved stoves program, which has reached more than
half the nation's rural households (>100 million stoves), only in recent years has the
percentage of success been high for such programs (Barnes et al„ 1992).
Recently, rising concerns about global warming from the buildup of CO2,
methane (CH4), and other greenhouse gases in the atmosphere have focused attention
on worldwide biomass combustion. Emitting 2100-4700 Tg carbon/y compared to S700
Tg C/y from fossil fuels, biomass burning plays important roles in the global carbon cycle
(Omtzen & Andxeae, 1990). Approaching 1000 Tg C/y, household biofuel, in turn.
4.
5-62
-------
accounts for a significant fraction of overall biomass combustion (Meyers & Leach,
1989). A question is thus raised: "Would alterations in household biomass combustion,
such as might be brought about by improved stoves, have significant implications for
global waiming?"
This question can be divided into three pans:
1. At the global (macro) scale, what are the contributions of biomass-buming
cookstoves in less-developed countries (here abbreviated BC-LDC) to global
inventories of major greenhouse-related emissions?
2. At the project (micro) scale, what are the technical and economic potentials for
reducing greenhouse-related emissions by changing BC-LDC technologies?
3. At the policy (meso) scale, what are the health, energy, and global wanning
implications of various policies affecting BC-LDC?
THE PILOT STUDY
To explore these issues, it is essential to know the BC-LDC emission factors of all
airborne species that have significant implications for energy, health, and global
wanning. Some of this information is already available, for in the 1980s a number of
studies were undertaken to examine the energy efficiency and health implications of
biomass stoves, both in developed and developing country situations (Smith, 1987).
Although a significant amount of greenhouse-gas research has gone into studies of
developing-country biomass burning at large scale (forest fires, swidden agriculture,
savannah burning, etc.), however, relatively little attention has focused on the type of
small-scale combustion found in BC-LDC (Levine, Z991).
Before attempting to fill this gap by embarking on a large-scale study of BC-LDC
greenhouse-related emissions, we and our colleagues decided to conduct first a
relatively small pilot study, litis could serve the double purpose of:
a. exploring whether the results were of sufficient interest to warrant
conducting a larger study; and
b. field testing some of the sampling and analysis techniques that might be used
in the larger study.
With these goals in mind, a small study of cookstoves in Manila was undertaken.
Monitored were emissions of more than 80 greenhouse-related and health-related gases
(mostly non-methane hydrocarbons, NMHCs) from traditional cookstoves burning
wood, charcoal, kerosene, and liquefied petroleum gas (LPG), which together account
for the majority of all cooking in developing countries. Involving only a few stove/fuel
combinations in each category, it is not possible to draw statistically valid global
conclusions from this pilot study. Nevertheless, the measurements are quite suggestive,
illustrating how more detailed studies of this type could be useful in answering the three
questions posed above.
5-63
-------
Table 1 summarizes the results of the sampling and analyses in terms of emission
factors, grams of pollutant per kilogram of fuel for each of the major PICs*. The few data
points argue that these results must be seen as tentative, until they can be verified at
larger scale. Fuels are presented in order of increasing health effects and decreasing fuel
carbon content.
TABLE 1. EMISSION FACTORS, GRAMS PER KILOGRAM DRY FUEL
n
Fuel
Carbon
Content
°°2
OO
CH,
TNMOC
TSP
LPG
2
0.87
3190
25
0.01
3
0.10
Kerosene
7
0.86
3050
39
0.90
14
3.00
Charcoal
6
0.80
2570
210
7,80
4
1.70
Wood
9
0.50
1620
99
9.00
12
2.00
n - Number of data points.
TNMOC (Total Non-Methane Organic Compounds): Per carbon molecular weight
taken as 18.
TSP (Total Suspended Particulates): Considered 75% carbon, see footnote 3.
EVALUATION
A useful way to evaluate both the local and global effects of BC-LDC is to detail
their impact on the carbon cycle. Shown as the framework to do so is the caibon flow
common to Figures 1-3, which is derived for the composite wood-fired cookstoves in the
Manila study. It follows the typical fate of the 500 g of carbon contained in 1.0 kg of
wood burned in such stoves. About 88% of the carbon is emitted as CO2 (weighing 1.6
kg) and the rest (60 g) is distributed as shown in Figures 1-3 among several kinds of
PICs, which together weigh about 126 g.3
Such a framework allows an examination of this flow in the context of the three
most important aspects of the emissions: energy, health, and global wanning.
These were determined by taking field samples from the flue stream of each above into stainless
steel canisters that were sent back to the U.S. far laboratory analysis. The ratio of each PIC to
CO2 (net of background levels) was used to determine emission factors by taking literature values
for the carbon contents of each fuel type and constructing the caibon balance for each stove.
Details are found in Smith et ai, 1992b.
3Paniculates were not measured in the pilot study, so data from other studies were used in the
figures (Smith, 1987; Joshi et aLv 1989; Smith, 1990).
5-64
-------
The Cookstove Energy Cycle
M% Cflmbuatiofl iHlctaney
22 kJ/c 174 mj
it M
1 kg Wood
SOOg Carbon
I I T 1J
CO CH< TNHQC R8f»
Energy Factor (ki/c) ¦ a< 74 78 "
_ i* a* eu ao7
Figure 1. This shows the movement of fuel carbon through a traditional wood-fired
cookstove as measured in the Philippines (Smith et aL, 1992a&b). Sixty grams of
carbon was not combusted completely; i.e., was released as PICs. Based on the available
energy in each PIC, if it had all been combusted completely, another 2J2 MJ would have
been released as heat. The stove, therefore, has a combustion efficiency of about 89%.
All the numbers refer to grams of carbon alone; e.g., the full mass of CO would be 28/12
(2.33) times larger. Here, total non-methane organic compounds (TNMOCs) are used
instead of NMHCs. Respirable suspended particulates (RSPs) are used instead of TSP.
Greater than 90% of TSPs are RSPs. NMHCs are about 94% of TNMOCs in this stove.
Source of energy contents: Lelieveld & Cnitzen (1992).
5-65
-------
Energy - To put the PICs in an energy context, each constituent needs to be weighted
by its energy content; i.e., the additional energy that could have been released if it had
been burned all the way to CO2. As shown in Figure 1, the result is that the PICs
contain about 11% of the energy originally in the wood; i.e„ the combustion efficiency is
about 89%. In other words, compared to a stove with near 100% combustion efficiency,
this stove requires about 12% more fuel (1/0.89).
This inefficiency is pan of the reason that traditional stoves use more fuel than it
seems they should. The other major technical reason, of course, is low heat transfer
efficiency (the fraction of heat released from the fuel that is taken into the cooking
utensil).
Health - As well as representing an energy loss, the 126g of PICs represent the main
health-damaging air pollutants from wood combustion. One way they can be
aggregated and compared is by use of the Relative Hazard Index (RHI). This is simply
the amount of air it would take to sufficiently dilute each pollutant until it reached the
relevant health-based concentration standard (Smith, 1987). With standards as shown,
the total RHI of the PICs is about 120,000 m3. CO2 is not much of a health hazard, as
shown by the relatively small RHI, 1800 m3. (Obviously, application of different
standards (e.g., from different countries) would result in different weightings for the
pollutants.)
A practical use of these dilution volumes would be to estimate what fraction of
the needed dilution might actually be achieved in a typical cooking situation. If, to be
generous, a village kitchen is 40 m3 and its air exchange rate is equivalent to about 25
air changes per hour (Smith, 1987), the kitchen has access to about 1000 m3 of dilution
air each hour. A typical burn rate for a woodstove is about 1 kg/h, producing each hour
the amount of PICs shown in Figure 2. Even in this fairly well-ventilated situation, the
available dilution air would seem to be far below (40-70 times) what is needed to keep
total non-methane organic compounds (TNMOCs)4 and respirable suspended particulate
(RSP) concentrations from exceeding the standards (the latter being more important for
health), a prediction consistent with many village measurements in developing countries.
In fact, it is not uncommon for indoor concentrations to reach 100 times the standard for
RSP during cooking (Pandey et alM 1989; Smith & Ahuja, 1990).
Two of the PIC categories shown (TNMOC and RSP) are composites containing
a vast array of mostly organic chemicals. Many of these individually are known to be
health-threatening (e.g., benzene in TNMOC and polyaromatic hydrocarbons in RSP).
Thus, if RHIs were calculated for each in turn, the total would be much larger than the
RHIs for the general categories.
'Strictly speaking, TNMOC is the appropriate category for following cartxm flows and NMHC for
global wanning implications. Since they are both about equal in size (NMHC are about 93% of
TNMOCs in this stove), however, we have not tried to maintain the distinction throughout the text.
5-66
-------
The Cookstove Health Cycle
120.000 mVc
CO
BO
PIC
CO
CM,
TNMOC RSP
Dilution Factor (mfa) :
0.1
MOO
30.000
10,000
70.000 40,000
1 kg Wood
500q Carbon
Figure 2. Starting with the same fuel carbon Hows as Figure 1, this figure weights the
PICs not on the basis of energy, but on the basis of how many cubic meters of air would
be necessary to dilute the emission to meet U.S. air pollution standards. Where there is
only an occupational standard, an appropriate safety factor (10) has been used to
establish a public standard. The following standards were used (in mg/m3): 002*900;
CO=10; CH4=11,000 (asphyxiation); NMHC=0.16; RSP^O.OS. The diluation factors
shown in the figure are on a pcr-carbon-atom basis. Although the NMHC standard was
actually set to prevent the formation of ozone, it represents a much less stringent
standard than would be applied if certain individual hydrocarbons were used as the
basis of the dilution factor. The benzene in NMHC from these woodstoves, for example,
would require 40 times more dilution than shown.
5-67
-------
Global Waiminp - Figure 3 evaluates the same PICs in terms of their greenhouse gas
potential. To do this, it is necessary to apply some index so that the impacts of the
different gases can be combined (Smith & Ahuja, 1990). This is so because the gases
have different heat-trapping abilities, lifetimes, and interactions with other gases in the
atmosphere. Here, we have used the Global Warming Potentials (GWPs) developed by
the Intergovernmental Panel on Climate Change (IPCC, 1990; Smith et ah, 1991).
These are given as a ratio to CO2 (either per molecule or per carbon atom), and thus
can be interpreted as the degree to which the total wanning of each compares to C02-
Since the gases have different atmospheric lifetimes, the relative impact (GWP) depends
on the chosen time horizon. Shown here are the results for time horizons of 20 and
100 years. In general, shorter time horizons make the non-CC>2 gases look more
important relative to CO2, since CO2 is the longest lived of this group.5
The result is that, depending on the time horizon chosen, the non-C02 GGs (i.e.,
the PICs) have a total GWP 20-110% as much as the CO2 itself. This implies that
looking only at the CO2 emissions of cookstoves may not give a good picture of their
global warming implications. It also implies that improvements in combustion
efficiency could result in much larger reductions in total GWP than would be indicated
simply by changes in CO2 emissions.
Biomass-Stove PICs - Relative Weights. From all three perspectives, energy, health,
and global warming, PICs are to be avoided. As shown in Table 2, however, the three
perspectives do not weight the individual PICs in the same way relative to one another.
Note that the weights are much more skewed for the health column than the others; i.e.,
a factor of 41 between NMOC and CO, down by an additional factor of 230 to the
minor hazard of CO2, and then another factor of 10 down to the insignificant health
hazard of CH4. Both the energy and global wanning viewpoints, in contrast, hold CH4
and TNMOC to have similar relative weights, and none of the differences are as large
as for health.
GLOBAL IMPACTS
Using these preliminary data, it is instructive to note how large BC-LDCs might
appear to loom in the global picture for each perspective.
SThese GWPs are not known with certainty and changes can be expected as knowledge improves.
Indeed, in its 1992 supplement, the IPCC (IPCC, 1992) suggested that indirect effects (chemical
interactions affecting other greenhouse gases) of the non-CXh gases were not well enough known
to be used in policy discussions; i.e., the values for the CH4 GWP would decrease to 13,4, and
13 (by time horizon) and those for CO and NMHC would be 1.0 at all times. The report states,
however, that "(t)he carbon cycle model used in these calculations probably underestimates both
the direct and indirect GWP values for all non-OCh gases." Given this caveat and that our purpose
here is principally illustrative, we have not modified the GWPs from those recommended originally
by the IPCC (IPCC, 1990). We have, however, used the updated estimates for the total effects of
CH4 as presented by Lelieveld & Crutzen (1992).
5-68
-------
The Cookstove Carbon Cycle
GWP (CO, Eoulvalant/c)
470
i l
170
t
440
A
1 kg Wood
5000 Carbon
440
Aah
n
0
M
T
•0
MC
I
1 43.8
¦
- 7
1J
MJEun
CO
CH,
NMHC
RSP
QWP (COj Equtvatont/c)
4M
22
12
1
1J
lMJCMra
QWP (COj Equt*ai*nt/C)
l.t
7.«
4.1
1
•1
M
14
Figure 3. The same carbon balance for the woodstove is shown as in Figures 1 and 2.
In this case, the PICs are weighted by the Global Warming Potentials (GWPs)
appropriate for 20-year and 100-year time horizons. Note that the PIC GWP is about
equal to that of the CO2 for a 20-year time horizon. Sources: Smith et al. (1992a&b);
Smith et al. (1991); IPCC (1992); Leiieveld & Crutzen (1992); Joshi et al. (1989).
5-69
-------
TABLE 2. RELATIVE IMPORTANCE FOR PJCs
UNDER DIFFERENT REGIMES
Global Wanning index
Time Horizon (years)
Energy
Health
20
100
500
CXI 2
1.0
1
1.0
1.0
1.0
GO
0.8
230
4.5
1.9
1.3
ch4
1.6
0
22.0
7.5
3.2
NMHC
1.8
9400
12.0
4.1
2.3
RSP
1.3
30000
1.0
1.0
1.0
All values are shown relative to CO2 on a carbon basis; i.e.t NMHC
is 9,400 times worse than CO2 for health.
Sources: GWPs from IPCC (1990); Smith et al. (1991) as corrected
by Lelieveld & Crutzen (1992). Energy data from Culp (1979).
Energy
Although humans in some way utilize perhaps 40% of world net biomass
production (Vitousek et al., 1986), as shown in Table 3, that proportion used directly
for fuel accounts for only about 15% of direct human energy use. Even so, biofuel in
the form of wood, crop residues, brush, and animal dung, is today still the chief form
of energy for most humanity, just as it has been since the discovery of fire (Hall &
Rosiilo-Calle, 1991). In developing countries, biofuels constitute about 35% of total
energy use, and in rural areas of developing countries, some 75%. In the poorest
developing countries, however, biomass fuels make up 80-90% of all energy use
(Smith, 1987). Based on the pilot study, therefore, the loss of energy represented by
the PICs from BC-LDC is roughly 1% of total human energy use and could approach
10% for some countries.
Health
In the case of health, particulate exposures from biomass use could be
responsible for approximately 50% of the total global human exposure. Most of this
occurs indoors in rural areas of developing countries, although there are significant
exposures in cities and outdoors as well. The vast preponderance of research,
regulation, and control of particulate air pollution is still focused on urban outdoor
developed-country situations, which, however, account for rather a small overall
fraction of global exposures (Smith, 1988).
5-70
-------
TABLE 3. GLOBAL IMPORTANCE OF PIC FROM B10MASS-F1RED COOKSTOVES
Energy
Biomass makes up about 14% of all direct human energy use.
h is about 33% of energy use in developing countries.
It is about 75% of energy use in niraJ areas of developing countries.
It is the most important fuel for the majority of humanity.
Sources: Smith (1987); Meyers & Leach (1989); Hall & Rosillo-Calle (1991)
Health
Cause of up to 50% of total human exposure to RSP.
Second largest occupational group, after farm workers (cooks).
Known risk factor for most important killer of developing-country children (pneumonia).
Sources: Pandey et al. (1989); Smith (1988)
Global Wurminp
Human biofuel consumption: 20-40% of all biomass combustion.
1-5% of all CH< emissions.
6-14% of all CO emissions.
8-24% of all NMHC emissions.
1-3% of all human-generated global warming.
Sources: Smith etal. (1992b); Ahuja (1990)
These high exposure levels are due not only to high particulate concentrations,
but also to the large populations involved. Indeed, after farmworkers, cooks represent
the largest occupational group in the world.®
It is important to note that the emissions from biomass fuels need not be high
compared, for example, to those from coal-fired industrial and power facilities in order for
the human exposures to be substantially greater. This is because a much larger
proportion of pollution released in household reaches people, compared to that from
centralized facilities. The impact per unit emissions tends to be greater for distributed
releases, and few things are more distributed than cooking, which occurs in every
household, every day.
Greenhouse Gases - Based on the few measurements taken in Manila, it would seem
possible that biomass stoves could account for fairly significant proportions of global
emissions of the three greenhouse gas categories — CO, CH«,and TNMOC (Table 3). For
CThese stoves arc undoubtedly responsible for a large fraction of global exposures to a range of
other pollutants as well; e.g., CO, polycyclic aromatic hydrocarbons, formaldehyde, and benzene.
5-71
-------
CO and CH4, the percentages in Table 3 translate into contributions to overall global
wanning from biomass-fired cookstoves of 0.4-0.9% and 0.1-0.5%, respectively. These
axe in same range as estimated by Ahuja (1990) for all biomass stoves, who also estimates
that the overall contribution of biomass stoves to global warming is about 2%. In
addition the contribution due to PICs is estimated to account for about 15% of net
deforestation and, thus, about 1.5% of net human CO2 additions to the atmosphere
(1.1% of total warming).
CONTROL MEASURES
Assuming people will continue to need to cook and are well versed in operating
their stoves, there are basically two ways to reduce PIC emissions from biomass-fired
cookstoves: change the fuel or change the stove. With new information of the kind
made available by the Manila study, it is possible to make further judgments about these
options.
Fuel
One objective of household energy policy can be to encourage people to move
up the energy ladder sooner than they otherwise might. This can be done through fuel
and stove pricing or other ways to make new stove/fuel combinations relatively more
attractive. In most parts of the developing world, the first step beyond unprocessed
biomass is charcoal or kerosene, followed by LPG. In some areas, Thailand, for example,
little kerosene is provided and the first step after charcoal is LPG. In China, it is often
coal, followed by LPG. Movement up the ladder generally results in substantially fewer
health-damaging PIC (RSP and CO) emissions per meal (Smith, 1990).
With a switch from biomass to fossil fuels, however, a global warming penalty
might at first seem inevitable because fossil rather than contemporary carbon would be
emitted. Because biomass combustion leads to a high amount of PICs with a large GWP
(Figure 3), however, the picture is substantially more complicated.
Based on the pilot study results, consider the benefits of switching from wood to
charcoal, kerosene, or LPG as summarized in Table 4.7 From a health standpoint, a shift
from wood to LPG reduces the overall health impact by a factor of 100. Kerosene, on
the other hand, results in a reduction by a factor of six. Use of a charcoal stove results in
an improvement by more than a factor of four.
From a global warming standpoint, the impact depends on how the biomass is
harvested; i.e., whether the CO2 is recycled. Table 4 shows the range between the
extremes; i.e., assuming totally sustainable harvesting and complete deforestation (no
regrowth). It is perhaps not surprising that, with complete deforestation, LPG and
kerosene are better than wood, 9 and 6 times, respectively. Even with sustainable
harvesting, however, because of the significant amount of PICs released by the
woodstove, kerosene and LPG stoves release, respectively, 2.6 and 1.7 times less GWP
pa- meal cooked.
7The values in Table 4 have been derived by taking into account the differences among the
stove/fuel combinations in overall cooking efficiency and energy per fuel carbon atom.
5-72
-------
TABLE 4. RELATIVE PIC EMISSIONS, HEALTH EFFECTS, AND GLOBAL POTENTIALS OF LPG,
KEROSENE. CHARCOAL, IMPROVED BIO MASS. AND WOOD COOKSTOVES
20-YEAR WARMING
Stove
Eff.
Biofuel
Use
CO
CH4
TNMOC
RSP
Health
Deforest
Regrow
LPG
0.70
0.00
0.02
9J0E-05
0.02
0.01
0.01
0.11
0.39
Kerosene
0.50
0.00
0.05
0.01
0.13
0.17
0.16
0.17
0.61
Charcoal
0.30
1.60
0.73
0.30
0.10
0.25
0.22
0.91
0.29
roc
-
wood
0.25
1.60
0.20
0.20
0.20
0.20
0.20
0.20
0.20
outdoor
0.40
0.40
0.40
0.40
0.04
0.40
0.40
Wood
0.15
1.00
1.00
1.00
1.00
1.00
1.00
• 1.00
1.00
Note: Except for the first, all the columns are normalized such that the impact of the traditional wood-fired
cookstoves is set to 1.0. Where appropriate, the impacts of the improved wood-fired cookstove with
flue are divided between those that occur indoors and outdoors. Elimination of indirect effects for CO
and TNMOC (NMHQ as suggested by IPCC (1992) changes the relative magnitudes, but does not
change the conclusions that these fossil-fuel cookstoves produce less GWP than this woodstove even
with renewable harvesting.
These values are somewhat misleading, however, because they exclude PIC
contributions from elsewhere in the fuel cycles for these fuels. The releases at oil fields
and refineries for kerosene and LPG, however, are likely to be less than 10% of those at
stoves (Smith et al., 1975). For locally harvested wood, they should be even lower,
although there may be some wastage during harvesting, transport, and storage.
For charcoal manufacture, however, the contribution is likely to be quite large.
Although there are few emissions data from the charcoal kilns commonly used in
developing countries, even relatively modern kilns apparendy emit rather large amounts
of PICs (Foley, 1986). Based on an extrapolation from emissions measurements of a U.S.
Missouri kiln (USEPA, 1986) operating at 33% efficiency (mass charcoal/dry wood).
Figure 4 shows the possible carbon flow in a kiln operating at 20% efficiency, which is
common in developing countries (Foley, 1986; Katyega and Kjeilstrom, 1991). Note that
nearly 35% of the carbon in the wood put in the kiln is released as PICs. Thus, the
20-year GWP ratio of PICS/CO2 for the kiln is about 7.6; i.e., the PICs produce more than
seven times as much GWP as the CO2 (4.8 times the 100-year GWP).
If the 64 million tonnes of wood made into charcoal each year in developing
countries (Joshi et al., 1989) was charged to kilns like the one in Figure 4, charcoal
making might be responsible for releasing 1.5 Tg/yr of carbon as CH*. This source may
be significant but direct measurements would help develop reliable estimates for the
types of the kilns typically in use.
5-73
-------
Wood
Wat Wood Water
Wood
6.0
1.0
6.0
Carbon
2.6
KHn
1.0
0.8
t
Charcoal
0.9
0.8
PIC
0.19 - CO
0.11 - CH<
0.27 - TNMOC
0.24 - TSP
Figure 4. Carbon flows for hypothetical charcoal kiln used in developing countries.
Based on measurements reported in U.S. EPA's Compilation of Air Pollutant Emission
Factors, 1986 (U.S. EPA, 1986) for a Missouri kiln with 33% production efficiency. This
has been modified to a 20% efficient kiln, which is more typical in developing countries,
by assuming that the difference in efficiency does not alter the ratio of Plus to CO2.
TOMOC=total non-methane organic compounds (vapor); TSP=total suspended
particulates (aerosol).
5-74
-------
100-
90-
80-
!
».
O)
50-
40-
30-
20-
10-
Combustlon
H(tt tr«n«Ur
Overall
Traditional IC-1
Stove Type
—r—
IC-2
Figure 5. The differences among overall and internal efficiencies in three metal wood-
fired cookstoves without flues. Note that, although both improved stoves achieve
substantially more overall energy efficiency than the traditional stove, combustion
efficiencies are less. Thus, IC-2 produces 4 times more PICs per unit energy delivered
than the traditional stove (100-92)/(100-98). Thus, though its energy use per meal is 2JS
times less (15/40), the overall result is that about 60% more PICs are produced per meal.
The original investigators measured only CO and particulates (Joshi et al., 1989). The
remaining PICs have been assumed to appear in the same ratios as measured in the
Manila pilot study.
5-75
-------
The first-level approach to improved biomass cookstoves (IBCs) is shown in
Table 4. The IBC uses 40% less fuel and releases about 359b of its PICs into the kitchen,
the rest going outdoors where it is assumed to have only 10% as much health impact per
gram as emissions released indoors. (These changes are based on results found in field
measurements of improved stove programs (Pandey et al., 1990; Ramakrishna et al.,
1989; Reid et al., 1986). Improvements measured in the laboratory can be much greater.)
Note that, due to lack of data, there is no change considered in global wanning impact
other than that 40% less fuel is used; i.e., the same fraction of the carbon is released as
PICs of the same composiuon.
To understand how changes in stove design and operation actually affect PIC
emissions, it is important to recognize that overall stove efficiency (Et) is a function of
two internal efficiencies -- combustion efficiency (Ec) (i.e., the amount of chemical
energy in the fuel that is convened to heat) and heat-transfer efficiency (Eh) (i.e., the
amount of heat that reaches the food in a cooking stove or reaches the room in a heating
stove):
Et = (Ec) x (Eh) (1)
In general, emissions per meal of PICs and CO2 are an inverse function of overall
efficiency in that, all else being equal, the less fuel used for a given cooking task, the less
PICs will be released. Thus, improvements in fuel efficiency should lead to lower
greenhouse gas (GHG) emissions.
Changes in stove operation and design, however, often affect the two internal
efficiencies in quite different ways. In particular, thermal transfer efficiency can be
increased at the expense of combustion efficiency. Design and operation changes that
improve overall fuel utilization, therefore, sometimes actually increase one internal
efficiency at the expense of the other.
Although few data axe available for biomass cookstoves, Figure 5 illustrates this
effect in a study of particulate and CO emissions of one traditional and two improved
wood-fired metal cookstoves (Pandey et al., 1989). Overall efficiency rose from 15% to
31% and 37% in the two improved stoves, greatly decreasing potential fuel demand for
cooking. In the process, however, PIC emissions per meal actually increased by 8%
because combustion efficiency dropped from 97% to 92%.
It might be thought that there is little net GHG impact from changes in
combustion efficiency. In other words, the fuel carbon that is not oxidized all the way
to CO2 will be released as PICs. The smaller the fraction of carbon released as CO2. the
more as PICs and vice versa.
In rough terms, this trade-off is true for carbon mass and number of carbon atoms.
It may not be true for the net greenhouse impact, however, because these different
molecules have different greenhouse impacts. Thus it is necessary to keep track not
only of the total carbon emissions but also of their fonn.
5-76
-------
The PICS/CO2 ratio can vary dramatically even at constant overall efficiency,
depending on the relative contribution of heat transfer and combustion efficiencies.
CO2 emissions are in general less dependent than PICs on combustion efficiency. For
example, a shift from 90% to 80% combustion efficiency results in a near doubling of
PICs but only about 10% less CO2. (More dramatically, a change in combustion
efficiency from 99% to 98% would result in less than a 1% loss of total efficiency but a
near doubling of PICs.)
Since PICs emissions are a stronger function of combustion efficiency than they
are of total efficiency, emissions can sometimes increase along with total efficiency. A
popular means by which fuel utilization of traditional cookstoves has been Taised,
regrettably, is simply to reduce airflow by enclosing the fire, thereby greatly increasing
the heat transfer efficiency to the pot, but also lowering the combustion efficiency. The
end result, therefore, can be a net increase in fuel utilization and a consequent reduction
in CO2 emissions, but a rise in the PICS/CO2 ratio or even an increase in absolute PIC
emissions per cooking task, as shown in Figure 5.
The GHG implications of stove emissions depend strongly not only on the
PICS/CO2 ratio, of course, but also on the particular mixture of PIC molecules. Each
mixture will have a different greenhouse equivalence weighting depending on the
relative amounts of the different constituents.
Hie radiatively active PIC molecules, such as CH4. and the molecules that play a
part in their atmospheric chemistry before turning into CO2, such as CO and NMHC,
have total (direct and indirect) GWPs above 1.0; i.e., greater than CO2. Indeed, it would
seem that all organic molecules must have a GWP per carbon atom of at least 1.0 because
once released they would relatively soon be oxidized to CO2, the rating of which, by
definition, is 1.0. Only elemental carbon panicles might have a GWP less than 1.0, if they
are assumed not to be oxidized within a relevant time period.
Although there do not seem to be sufficient theory or data to predict the exact
relationship between design changes and the PICS/CO2 ratio, we do have rough
estimates of typical GWPs of PICs. The PIC GWP of the woodstoves in Manila varied
from 1.7 to 7.8, depending on the time horizon.
Thus, efficiency improvements to the Manila woodstove that allowed combustion
efficiency to drop in exchange for increased heat-transfer efficiency could actually lead
to significant increases in PICs with their health and greenhouse impacts.
Final decisions with regard to fuel and stove changes will of course depend on
relative economics and other non-environmental issues (Smith, 1992). Nevertheless, the
carbon flow framework made possible by the monitoring data would be a valuable
grounding for these further analyses.
CONCLUSION
Putting aside for the moment the few actual measurements involved, the
infonnation available from the Manila pilot study has allowed us to make substantial
progress toward answering the three questions posed at the beginning:
5-77
-------
!. The contribution to global inventories;
2. Advantages and disadvantages of various technical options (stove/fuel
combinations); and
3. Policy implications because of interactions among energy, health, and global
wanning objectives.
As we indicate above, it would be extremely useful to have sufficient data to
resolve the carbon balances of such small combustion devices as stoves. Five
characteristics make such devices attractive for this kind of research:
a. The devices are (or are operated in a way to be) substantially different
from those in developed countries, which have been subject to much
research already;
b. They have widely varying, but generally poor, combustion efficiencies,
leading to significant amounts of PICs;
c. Although individually small, they are widely used, leading potentially to
emissions significant on the global scale;
d. There is substantial scope for technical improvement; and
e. Of interest, although not of direct concern from the standpoint of global
warming, they are evenly dispersed with the population, making their PIC
emissions more likely to produce ill health.
Stoves fill this bill, as do other ubiquitous combustion devices such as motor vehicles.
Based on these encouraging, but preliminary findings, we are planning to embark
on a more extensive study of cooking and heating stoves. This will be undertaken
jointly with colleagues in India and China, which not only contain about 65% of the
population in developing countries, but where a wide range of stove/fuel combinations
are in use. Because of the potentially, but little studied, significance of charcoal kiln
emissions, we also plan to study the emissions of a wide range of kilns in Thailand and
Brazil.
We also hope to examine more closely a set of greenhouse-gas sources in
developing countries that have potential for rather large contributions to global
inventories. These are garbage dumps around large cities where not only is there
significant anaerobic decay, but also smoldering spontaneous combustion, low-tech
incineration, and intentional burning by scavenger communities to recover metals.
5-78
-------
REFERENCES
Ahuja, D.R., 1990, "Research Need for Improving Biofuel Burning Cookstove
Technologies," Natural Resources Forum 14(2): 125-134.
Baldwin, S., 1987, Biomass Stoves, Volunteers in Technical Assistance, Princeton
University, Arlington, VA.
Barnes, D.F., et al., 1992, "What Makes People Cook with Improved Stoves? A
Comparative Review," ESMAP/UNDP Paper, World Sank, Washington, D.C.
Crotzen, PJ. & M.O. Andreae, 1990, "Biomass Burning in the Tropics: Impact on
Atmospheric Chemistry and Biochemical Cycles," Science 250:1169-1678.
Culp, A.W., 1979, "Principles of Energy Conversion," McGraw Hill, New York, NY.
Foley, G., 1986, Charcoal Making in Developing Countries, Earthscan, London.
Hall, D.O. & F. Rosillo-Calle, 1991, Biomass in Developing Countries, Report to the
Office of Technology Assessment, Washington, D.C.
IPCC (Intergovernmental Panel on Climate Change), 1990, Climate Change: The IPCC
Scientific Assessment, Cambridge University Press, UK.
IPCC, 1992, Supplement, Cambridge University Press, UK.
Joshi, V. et al., 1989, "Emissions from Burning Biofuels in Metal Cookstoves,"
Environmental Management 13(6): 763-772.
Katyega, MJ., & B. Kjeilstrom, 1991, "Assessment of Forest Biomass Technology," ATAS
Bulletin, #6,139-148.
Lelieveld, J. & PJ. Crotzen, 1992, "Indirect Chemical Effects of Methane on Climate
Wanning," Nature, Vol. 355:339-341.
Levine, J.S., ed., 1991, Global Biomass Burning, MIT Press, Cambridge, MA.
Meyers S. & G. Leach, 1989, "Biomass Fuels in the Developing Countries: An
Overview," LBL-27222, Lawrence Berkeley Laboratory, Berkeley, CA.
Pandey, M.R., et al., 1989, "Indoor Air Pollution in Developing Countries and Acute
Respiratory Infections in Children," Lancet Feb.25:427-429.
Pandey, MH., et al., 1990, "The Effectiveness of Smokeless Stoves in Reducing Indoor
Air Pollution in a Rural Hill Region of Nepal," Mountain Research and Development,
10(4): 313-320.
Ramakrishna, J., et al., 1989, Cooking in India: The Impact of Improved Stoves on Indoor
Air Quality," Environment International 15(1-6): 341-352.
5-79
-------
Reid, HP., etal., 1986, "Indoor Smoke Exposures from Traditional and Improved
Cookstoves: Comparisons among Rural Nepali Women," Mountain Research and
Development, 6(4): 293-304.
Smith, K.R., et al., 1975, Evaluation of Conventional Power Systems, Jet Propulsion
Laboratory, Pasadena CA.
Smith, K.R., 1987, Biofuels, Air Pollution, and Health, Plenum Press, New York, NY.
Smith, BLR., 1988, 'Total Exposure Assessment: Pan 2, Implications for Developing
Countries," Environment 30(8): 10-15; 33-38.
Smith, K.R., 1990, "Indoor Air Quality and the Pollution Transition," in H. Kasuga, ed.,
Indoor Air Quality, Springer-Verlag, Berlin, 448-456.
Smith, K.R., et al., 1991, "Indices for a Greenhouse Control Regime That Incoiporates
Both Efficiency and Equity Goals," Working Paper 91-21, Environmental Policy and
Research Division, World Bank, Washington, D.C.
Smith, K.R., 1992, "Biomass Cookstoves in Global Perspective," in World Health
Organization, Indoor Air Pollution from Biomass Fuel, WHO/PEP/92.3B, Geneva, 164-
184.
Smith, K.R., et al., 1992a (In Press), "Greenhouse Gases from Biomass and Fossil Fuel
Stoves in Developing Countries: A Manila Pilot Study," Chemosphere.
Smith, K.R., et al., 1992b, "Greenhouse Gases from Small-Scale Combustion in
Developing Countries," EPA-600/R-92-005 (NTIS PB92-139369), Air and Energy
Engineering Research Laboratory, Research Triangle Park, NC.
Smith, K.R. & D,R. Ahuja, 1990, "Toward a Greenhouse Equivalence Index: The Total
Exposure Analogy," Climatic Change 17: 1-7.
USEPA, 1986, Compilation of Air Pollutant Emission Factors, AP-42 (NTIS PB87-
150959), Washington, D.C.
Vitousek, P., et al., 1986, "Human Appropriation of the Products of Photosynthesis,"
Bioscience, 36(6): 368-373.
5-80
-------
The Potential for Energy Crops to Reduce Carbon Dioxide Emissions
R. L. Graham
Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee 37801-6038
Energy crops are herbaceous or woody plants grown specifically to produce biomass for combustion
and production of electricity or for conversion to fuels such as ethanol. When grown intensively such
crops can yield up to 43 dry Mg biomass/ha/yr in the temperate zone although yields of 10 to 20 Mg
are more typical.
By substituting sustainably-grown biomass for fossil fuels, C02 emissions from energy consumption
can be reduced significantly. Although biomass fuels are net emitters of C02 because fossil fuels are
used in the production of energy crops (e.g., fertilizers, transportation), biomass fuels emit much less
C02 than fossil fuels per unit of energy produced. A hectare of U.S. farmland on the average could
produce sufficient biomass to reduce annual C02 emissions by 5 Mg C if such biomass was used to
produce electricity that would otherwise be produced with coal.
Within the U.S. there are at least 131 million hectares of farmland that could support energy crop
production. Much of this land lies in the midwest although the southeast also contains considerable
acreage. More acreage can support herbaceous energy crops than woody crops as there are
herbaceous crops that can tolerate drier conditions than any woody crop.
The potential for energy crops to reduce C02 emissions will depend on how much land can be
profitably dedicated to energy crops. Fanners will adopt energy crops when it is economically
advantageous for them to do so. Power companies will adopt biomass energy when the cost of energy
from fossil fuels exceeds the cost of energy Gram biomass. Government policies and regulations affect
both these relationships. Currently biomass crops must compete with subsidized agricultural crops
thereby inflating the price of biomass to an energy facility. On the other hand, pollution regulations
can favor energy crops as a fuel source since biomass crops can bum cleaner than coal
Research sponsored jointly by Office of Policy Analysis, Environmental Protection Agency (EPA),
under Interagency Agreement Number DW89934775-01 (DOE 1824-C092-A1) with the Biofuels
Systems Division, U.S. Department of Energy, under contract DE-AC05-840R21400 with Martin
Marietta Energy Systems, Inc.. This document has not been subjected to EPA review and therefore
does not necessarily reflect the views of EPA and no official endorsement should be inferred. See
following paper.
5-81
-------
(NOTICE: Although the following paper was not presented at the
Symposium, it was provided by principal author Graham as being
equivalent to the presentation made there and representative of the
related research.)
THE POTENTIAL FOR SHORT-ROTATION WOODY CROPS
TO REDUCE U.S. C02 EMISSIONS*
by: R, L. Graham1
L. L. Wright1
A. F. Turhollow2
Environmental Sciences Division1
Energy Division2
Oak Ridge National Laboratory
P.O. Box 2008
Oak Ridge, TN 37831-6352
"Research sponsored by the Biofuels Systems Division, U.S. Department of
Energy, under contract DE-AC05-840R21400 with Martin Marietta Energy Systems,
Inc.
5-82
-------
1
Abstract
Short-rotation woody crops (SRWC) could potentially displace fossil
fuels and thus mitigate C02 buildup in the atmosphere. To determine how much
fossil fuel SRWC might displace in the United States and what the associated
fossil carbon savings might be, a series of assumptions must be made. These
assumptions concern the net SRWC biomass yields per hectare (after losses);
the amount of suitable land dedicated to SRWC production; wood conversion
efficiencies to electricity or liquid fuels; the energy substitution
properties of various fuels; and the amount of fossil fuel used in growing,
harvesting, transporting, and converting SRWC biomass. Assuming the current
climate, present production, and conversion technologies and considering a
conservative estimate of the U.S. land base available for SRWC (14 x 106 ha),
we calculate that SRWC energy could displace 33.2 to 73.1 x 106 Mg of fossil
carbon releases, 3-6% of the current annual U.S. emissions. The carbon
mitigation potential per unit of land is larger with the substitution of SRWC
for coal-based electricity production than for the substitution of
SRWC-derived ethanol for gasoline. Assuming current climate, predicted
conversion technology advancements, an optimistic estimate of the U.S. land
base available for SRWC (28 x 106 ha), and an optimistic average estimate of
net SRWC yields (22.4 dry Mg/ha), we calculate that SRWC energy could displace
148 to 242 x 106 Mg of annual fossil fuel carbon releases. Under this
scenario, the carbon mitigation potential of SRWC-based electricity production
would be equivalent to about 4.4% of current global fossil fuel emissions and
20% of current U.S. fossil fuel emissions.
5-83
-------
Biomass energy is one of several renewable energy sources that could be
used to displace fossil fuel energy and thus reduce fossil fuel C02 emissions
to the atmosphere. Among the renewable energy technologies, biomass is unique
in that it can be a source of not only electricity but also liquid or gaseous
fuels. Thus biomass energy is the one renewable energy source that can
directly displace such petroleum products as gasoline and diesel fuel.
Renewable biomass feedstock can come from either herbaceous or woody
crops. In this paper we will focus on wood feedstocks and in particular wood
from tree crops grown under short rotations (4-12 years) in an agricultural
production system hereafter referred to as short-rotation woody crops (SRWC).
SRWC have been promoted as a possible source of significant amounts of
renewable woody feedstock and a technology that could offer landowners an
attractive alternative use (rather than food production) for marginal to good
cropland.
The objective of this paper is to estimate the potential for SRWC
technology to reduce the U.S. C02 emissions in view of both current and future
technologies. To make this estimation, a series of linked questions must be
addressed. First, how much SRWC energy feedstock could be produced annually
in the United States? Second, how much electricity or fuel could that
feedstock produce? Third, how much fossil fuel could that wood-derived
electricity or fuel displace? Fourth and finally, what would be the net
carbon mitigation benefit to the atmosphere of that displacement? The links
among these questions are outlined in the following equations.
5-84
-------
3
Amount of feedstock =
f1(number of hectares, yield/ha, harvest & storage losses)
Energy produced =
f2(amount of feedstock, conversion efficiency)
Fossil fuel carbon displaced -
f3(energy produced, fuel substitution properties, carbon contents of
fossil fuels)
Net carbon mitigation =
f4 (fossil fuel carbon displaced, carbon input to SRWC feedstock)
We will briefly discuss factors affecting feedstock amount, energy
production, fossil fuel displacement, and net carbon mitigation benefit. Then
preliminary estimates of the possible carbon mitigation benefit of growing
SRWC for energy production in the United States will be presented on the basis
of current and future technologies and two levels of land availability.
Amount of Feedstock
To determine the amount of wood feedstock that might be produced by
growing SRWC, it is necessary to know the wood yield per hectare of SRWC and
5-85
-------
4
the amount of land available and suitable for SRWC. Yields and land
availability will be addressed separately.
Yield Potentials
In considering SRWC yields, several questions must be posed. First,
what yields are currently and potentially attainable? Second, what yields are
necessary for SRUC energy to be competitive with other energy types? And
third, what technology and site conditions are needed to produce competitive
yields?
SRWC yields in research field trials have steadily risen over the past
12 years as SRWC technology has improved. Yields as harvestable woody biomass
in the range of 9-15.7 (dry) Mg/ha/year are now common in production research
trials (Wright e£ al. , 1992). (Yield is defined as "the standing aboveground
woody biomass of the SRWC crop divided by the crop's age" and is analogous to
the forestry term mean annual increment. or MAI.) If a concerted research
effort is maintained, yields of 15.7-29.1 (dry) Mg/ha/year are believed
attainable by the year 2010 on moderate to good cropland. To date, the
maximum observed experimental yield from a temperate climate SRWC is 43.3
(dry) Mg/ha/year (Table I) (Wright and Ehrenshaft, 1990).
To be competitive with other energy sources at current prices in the
U.S. market, delivered SRWC energy feedstock must cost less than $1.90/GJ
[$37/(dry) Mg] in U.S. dollars (Ranney et al., 1987). This would permit
biomass-to-ethanol production costs to match current gasoline production
costs. The cost rate translates to a yield of roughly 16-22 (dry) Mg/ha/year
on sites having moderate annual land rents ($100-$148/ha). Current delivered
5-86
-------
5
SRWC costs are roughly $1.90-$2.85/GJ [$37-$56/(dry) Mg] when relatively good
cropland is used, moisture is not limiting, and the best available SRWC
techniques are practiced (Wright and Ehrenshaft, 1990). These cost figures
assume that chips are used as the raw feedstock. If whole trees could be
burned as has been proposed (Ostlie, 1989), the feedstock costs would be
significantly less. The economic attractiveness of wood as an energy
feedstock is, of course, highly dependent on the cost of alternative energy
sources. Taxes on fossil-carbon-emitting sources or increases in oil import
prices would improve the competitiveness of SRWC as an energy source. However,
currently high SRWC production costs and low fossil fuel prices combine to
make SRWC-derived energy noncompetitive in most situations.
SRWC yields are a function of genotype, cultural practices, and site
quality. Silver maple (Acer saccharinura). sweetgum (Liauidarober
stvraciflua), American sycamore (Platanus occidentalis), black locust (Robinia
pseudoacacia). poplars (PopuIus spp. and hybrids), and eucalypts (Eucalyptus
spp.) have been identified as potential species for SRWC in the United States
and are currently the subject of SRWC research. Hybrid poplars and eucalypts
have shown the greatest potential thus far for attaining exceptionally fast
growth rates in the United States. Good cultural practices are critical if
promising genotypes are to express their growth potential. Site preparation
is especially important. SRWC must be established under the same conditions
as almost any other agricultural crop, and fertilization is needed for most
SRWC species approximately every other year to maintain rapid growth and
sustain fertility of the land. To obtain high SRWC yields at reasonable
costs, agricultural cropland is required [land in U.S. Department of
5-87
-------
6
Agriculture (USDA) cropland classes IV or better] (Wright ££ al., 1992). The
establishment of SRWC crops on cut-over forested sites has been tested, but
results have shown that growth was relatively poor compared with that of
cropland under cultivation or recently abandoned (Wright al., 1989). For
example, sweetgum and sycamore yields on an old-field site in coastal Alabama
were more than double those on an adjacent cut-over forested site. Thus, from
both an economic and a carbon mitigation perspective, the conversion of
forested land to SRWC is generally not desirable.
Harvesting and handling technology can affect both net yield from the
site and feedstock costs. Harvest losses are commonly assumed to be between 5
and 10% of the aboveground wood standing in the field at the time of harvest,
and storage losses are estimated at 10 to 15% (Wright et al., 1992).
Harvesting costs are difficult to project because harvesting equipment
suitable for the unique conditions of SRWC are under development and have not
been widely tested. . Smaller equipment than is commonly used in conventional
forestry is more cost effective, however (Stokes et al., 1986).
Land Availability
The land base that might be dedicated to growing SRWC in the future is a
function of land quality, agricultural product demands, U.S. agricultural
policy, and environmental considerations. Of these four factors, land quality
can be discussed with the most certainty. An analysis of the U.S. nonfederal
land base suggests that there are 159 million hectares of cropland or
potential cropland that could under current conditions support SRWC without
the use of irrigation (Wright e£ al., 1992). Ninety-one million of these
5-88
-------
7
hectares might support SRWC yields in excess of 11 Mg/ha/year. The average
yield on these lands was projected to be 14 dry Mg/ha/year.
Projections of land needed for current and future food production in the
United States suggest that putting existing U.S. cropland into energy-crop
production may be both feasible and economically desirable (USDA/SCS, 1990).
In 1982, 62 x 106 ha of the approximately 200 x 106 ha potentially suitable
for crop production were either fallow or used for pasture or range (USDA/SCS,
1987). In addition to the land normally used for range and pasture, varying
amounts of productive cropland are idled each year. For any 1 year during the
1980s, 4.5 x 10s to 32 x 106 ha of productive cropland were idle. Thus, a
significant portion of the U.S. crop base is not used for food production.
This underutilization of the U.S. cropland base has received government
attention. A report requested by the U.S. Secretary of Agriculture (New Farm
and Forest Products Task Force, 1987) proposed a national goal of "developing
and commercializing within 25 years, an array of new farm and forest products,
utilizing at least 150 million acres [60 x 106 ha] of productive capacity, to
meet market needs representing net new demand for agriculture and forestry
production." SRWC would appear to be such a "new product," assuming energy
markets develop for SRWC feedstocks. Such markets are, however, unlikely to
develop unless biomass feedstocks become more competitive with fossil
feedstocks.
Furthermore, current institutional factors would inhibit the conversion
of land to SRWC production even if energy markets did develop. Present U.S.
agricultural policies provide incentives to keep resources in traditional
agricultural practices, and lending institutions would require a stable and
5-89
-------
8
long-term assured market for energy crops before financing became available.
To encourage adoption of SRWC, agricultural producers would probably require
assurances that commodity program hectares on their farm would not be affected
and SRWC markets are stable and permanent. To produce under the current
institutional structure, the price for the SRWC feedstock would need to
incorporate significant incentives to overcome the institutional factors as
well as the cost of producing the SRWC feedstock.
Changes in U.S. farm policies are beginning to occur. In 1985 the Food
Security Act incorporated provisions of the type needed to begin establishment
of wood energy crops. The act initiated the Conservation Reserve Program
(CRP) , which targeted 16 x 106 ha of highly erodible land for long-term
retirement from food crop production. Land eligibility is limited to
predominantly highly erodible cropland, filter strips, certain woodland areas,
and fields having evidence of scour erosion. As of 1989, 12 x 106 ha had been
set aside under this program. SRWC grown on 9- to 12-year rotations would
appear to be suitable for some of these lands because the establishment of
SRWC accomplishes the goal of reducing soil erosion yet is also not as
permanent a land-use conversion as conventional forestry. SRWC is potentially
profitable today on this land base if government CRP subsidies are included
(Lothner et al., 1988).
Energy Production
To determine the amount of energy in the form of electricity or fuel
that can be produced from a hectare of land or a megagram of biomass
5-90
-------
feedstock, it is necessary to understand the status of conversion
technologies, particularly the efficiency of conversion. Today's biomass
conversion technologies tend to have relatively low energy-conversion
efficiencies, but considerable potential exists to improve the overall process
efficiencies and make these processes more economically favorable (DOE/BMWTD,
1988).
Electricity Production
Most conventional wood-to-electricity generating plants (10-50 MWe)
operating today have an energy-conversion efficiency of about 25%, largely
because of the high moisture content of the wood feedstock. However, the use
of low-moisture-content wood feedstocks with high-pressure and high-
temperature turbines and boiler systems could result in wood-to-electricity
conversion efficiencies of 33-35% (Gary Elliot, president, Fitchner USA, and
Dave Ostlie, president, Energy Performance Systems, personal communication,
Spring 1990). Supercritical pressure plant cycle efficiencies as high as 40%
have been demonstrated (Combustion Engineering, 1981). Larson and Williams
(1989) and Larson and Svenningsson (1990) have proposed the use of
aeroderivative gas turbines [steam-injected gas turbines (STIG) and
intercooled steam-injected gas turbines (ISTIG)] coupled with biomass
gasifiers to generate electricity. Efficiencies for STIG and ISTIG, using
coal as the feedstock, have been documented at 35.6 and 42.1%, respectively.
Because biomass should gasify more easily than coal, Larson and Svenningsson
(1990) feel that biomass should have at least as high an overall cycle
5-91
-------
efficiency as coal if the gasifier system is specifically designed for
biomass.
Ethanol Production
Using an ethanol conversion facility model developed at the National
Renewable Energy Laboratory, researchers found that a dry megagram of wood can
produce about 344 L of ethanol with a conversion-process efficiency of 41%
(Hiranan et al., 1991). The ethanol is produced from the hemicellulose and
cellulose in the wood. The lignin and unfermented carbohydrates of the wood
are burned to provide energy for the conversion process. The wood-to-ethanol
process differs from the more conventional corn-to-ethanol production in which
the starch in the corn grains converts to ethanol and the process energy is
generally provided by coal (Marland and Turhollow, 1991).
The ethanol conversion model also indicates that current wood-to-ethanol
technology would produce a surplus 184 kWh of electricity per dry megagram of
wood processed to ethanol. Actually, the process would generate 495 kWh of
electricity, but 311 kWh would be used in processing the feedstock. With
improvements in both feedstock composition and the conversion process (e.g.,
fermentation efficiencies and enzymes used for hydrolysis of cellulose), a dry
megagram of wood could produce about 503.3 L of ethanol and a surplus of about
101 kWh of electricity. This represents a wood-to-ethanol conversion-process
efficiency of 60%. The reduced electricity production under the future
technology analysis is a result of more efficient use of the wood in the
ethanol conversion process and, consequently, less wood residue for
combustion.
5-92
-------
Gross Fossil Fuel Carbon Displacement
The gross amount of fossil fuel carbon that could be displaced per unit
of wood feedstock is a function of fuel substitution properties, the amount of
energy or fuel produced per unit of wood, and the amount of carbon released
per unit of fossil fuel energy. Because fuel substitution properties vary
considerably between ethanol and coal, the two will be discussed separately.
Coal Displacement
The relative amount of electricity produced per dry Mg of wood or coal
feedstock is a function of the process-conversion efficiency and the relative
energy content of the two feedstocks. The energy contents of both wood and
coal per unit of delivered weight vary considerably. An average energy
content of wood is 19.8 GJ/(dry) Mg, whereas an average energy content of coal
is about 27.2 GJ/Mg as delivered (Turhollow and Perlack, 1991; Marland and
Turhollow, 1991). These values, together with the assumed conversion
efficiencies of wood and coal combustion, determine the relative raw energy
levels of each feedstock required to produce the equivalent kilowatt hours of
electricity. The raw energy levels of wood and coal required to produce a
kilowatt hour of electricity are obviously the same if the conversion
efficiencies are the same.
•Once the energy level of the coal equivalent (to the wood burned) is
determined, the calculation of carbon offset is straightforward because the
carbon content of coal per joule is nearly constant over a wide range of coal
types. That carbon level in coal is 24.12 kg/GJ if carbon released in the
5-93
-------
mining and transportation of coal is not considered and 24.65 kg/GJ if the
additional carbon emissions are considered (Marland, 1983). The gross fossil-
carbon offset for substituting wood for coal is therefore 488.1 kg C/Mg wood,
assuming equal coal- and wood-conversion efficiencies (19.8 GJ/Mg wood *
24.65 kg G/GJ coal). The carbon offset per unit wood is not sensitive to
improvements in process conversion efficiencies because it is assumed the same
improvements would occur for coal conversion.
Gasoline Displacement
The calculation of gross fossil fuel displacement in the wood-to-ethanol
fuel pathway includes consideration of both relative substitution rates of
ethanol for gasoline and any electricity produced or fossil carbon released in
operation of the conversion facility. Liquid fuels derived from biomass do
not necessarily substitute for conventional fuels on a one-to-one energy
content basis. Even though a liter of ethanol contains only about two-thirds
the energy content of a liter of gasoline, it can be burned more efficiently.
Therefore, a liter of ethanol can substitute for 0.8 L of gasoline (Lynd et
al.. 1991). Much of the energy in the wood used to produce ethanol is needed
to provide process energy and thus is not available to displace gasoline.
Consequently, it takes about two units of wood energy to displace one unit of
energy in gasoline. This is in contrast to coal displacement for electricity
generation in which a unit of wood energy is equivalent to a unit of coal
energy.
Because the carbon content of gasoline is 20.76 kg/GJ (0.723 kg/L) if
the energy costs of refinement and transportation are included (Marland and
5-94
-------
Turhollow, 1991) , the gross ethanol carbon offset per dry Mg of wood is
199 kg C under current conversion technology (i.e., 344 L ethanol/Mg wood *
0.8 L gasoline/L ethanol * 0.723 kg C/L gasoline — 199 kg C/Mg wood). The
gross offset is 291. kg C under future conversion technology. If one then
includes the carbon savings associated with the surplus electricity
production, the total carbon offset is 236 kg C/Mg wood for current technology
and 307 kg C/Mg wood for future technology. These values incorporate the
assumption that 75% of the electric substitution would be from fossil fuel and
the other 25% from renewables so the kilowatt-hour carbon savings is
0.2 kg C/kWh for current technology and 0.16 kg C/kWh for future technology
(Turhollow and Perlack, 1991).
Net Carbon Benefits
The calculation of net carbon benefits requires some assumptions of
yield and SRWC management practices because they will determine the amount of
fossil fuel used to produce a unit of SRWC-derived energy. For purposes of
this discussion and for assessing the U.S. potential for SRWC-derived energy,
we assume that current technology will produce average standing yields of
14 Mg dry wood/ha/year, future technology will produce average standing yields
of 22.4 Mg dry wood/ha/year, and in both scenarios there is a 5% harvesting
loss and a 15% (current) to 13% (future) transportation and storage loss. As
noted previously, 14 Mg/ha/year was the average yield calculated for all U.S.
lands that could support SRWC with yields greater than 11 Mg/ha/year. The
22.4 Mg/ha/year yield under future technology was chosen to reflect current
5-95
-------
expectations for the future (Wright et a!., 1992). Conversion efficiencies
must also be assumed to calculate the gross fossil fuel carbon offset. For
our current scenario, we assume electric production efficiencies of 33% for
both wood and coal and an ethanol conversion efficiency of 41% (344 L of
ethanol plus 184 kWh of electricity per megagram of wood). For our future
scenario, we assume electric production efficiencies of 42% for both wood and
coal and an ethanol-conversion efficiency of 60% (503 L of ethanol plus
101 kWh of electricity per megagram of wood).
The combustion of wood or ethanol (products of SRWC) does emit carbon
into the atmosphere, but these emissions are balanced by the carbon taken up
by the SRWC energy plantations, provided there are no soil carbon changes.
However, fossil-carbon inputs are currently required in the production,
harvesting, and transportation of SRWC feedstocks. Soil carbon changes are
difficult to predict because they strongly depend on the former land use and
SRWC management practices. If there was no irrigation and the former land use
was conventional crop production, some increase in soil carbon might occur in
SRWC production as a result of the reduction in tillage and the unharvested
root systems of SRWC. However, data to document the level of sequestering in
soil and roots are not yet available, the sequestering would not continue
indefinitely, and the stored carbon might be released if the site were to be
converted back to food crop production. Growing SRWC on land previously
forested or in pasture or irrigating SRWC would most likely cause a net loss
of soil carbon. For purposes of this analysis, soil carbon is assumed not to
change, and all carbon inputs to the production, harvesting, and
transportation of SRWC are assumed to be carbon emissions that reduce the
5-96
-------
benefit derived from fossil fuel displacement. The following discussion
briefly describes the types and levels of carbon inputs and calculates, by
conversion technology - current and future, electricity and ethanol - the net
benefit derived from displacement of fossil-carbon feedstocks with SRWC. A
detailed discussion of these carbon inputs can be found in Turhollow and
Perlack (1991).
Carbon Inputs to SRWC Feedstock Production
Carbon inputs to the production, harvest, and delivery of short-rotation
woody feedstocks are a function of expected management practices and expected
yield. Energy requirements for each operation (site preparation, weed
control, fertilizer, pesticides, harvest, transport, and storage) must be
estimated by fuel type (diesel as a proxy for all petroleum liquids, natural
gas, and electricity) and then converted to carbon equivalents. For our
analysis of the potential for SRWC production in the United States,
electricity is assumed to be generated 75% from coal and 25% from nonfossil
sources. Management practices are assumed the same for both scenarios and are
described in greater detail in Turhollow and Perlack (1991) in their
description of hybrid poplar production. Harvest is assumed to occur on a
6-year cycle, with two coppice harvests, for a stand life of 18 years.
Establishment of the plantation requires an application of a contact
herbicide and disking in the fall followed by a disking and application of
pre-emergent herbicide prior to spring planting. Herbicides are applied only
the first and second years of the stand life. Phosphate and potash are
applied the first year of each cycle, nitrogen is applied every other year,
5-97
-------
and insecticides and fungicides are applied on the average every 5th year in
the 6-year cycle. The pesticide-use estimates in these scenarios are probably
high. Actual pesticide use in operational SRWC poplar stands has been much
less (Don Rice, James River Company, personal communication, Spring 1991).
With an exception of harvest, transport, and storage operations, fossil-fuel
use in SRWC production is calculated on a per hectare basis and is insensitive
to yield. Harvest, transport, and storage operations are assumed to use
diesel fuel at a rate of 1.1 GJ/(dry) Mg of wood delivered to the conversion
facility; thus, diesel use is sensitive to yield.
For current technology, production of an average annual (after loss)
yield of 11.3 (dry) Kg/ha requires per hectare inputs of 270 L of diesel fuel,
77 m3 of natural gas, and 84 kWh of electricity, which results in carbon
emissions of about 0.29 Mg C/ha/year or 25.8 kg C/Mg of wood delivered to the
conversion facility. With future technology, production of an average annual
(after loss) biomass yield of 18.5 dry Mg/ha requires per hectare inputs of
401 L of diesel fuel, 77 m3 of natural gas, and 84 kWh of electricity, which
contributes carbon emissions of about 0.40 Mg/ha/year or 21.9 kg C/Mg wood.
The difference in diesel-fuel usage results from more wood per hectare being
harvested, transported, and stored.
Net Carbon Benefit from Conversion of Wood to Electricity or Ethanol
The net carbon mitigation benefit of using SRWC to produce electric
power and/or ethanol is simply the carbon displacement of the fuel
substitution minus the carbon emissions from SRWC production. Tables II and
III outline these calculations on a per hectare basis. By substituting wood
5-98
-------
for coal, the gross carbon offsec is about 5.51 and 9.03 Mg C/ha/year using
available and future technologies, respectively. After subtraction of the
carbon emissions associated with SRWC production, harvesting, and
transportation, the net offset of carbon is about 5.22 and 8.63 Mg C/ha under
available and future technologies, respectively. The net carbon offset by
conversion of SRWC to ethanol to be used for gasoline substitution amounts to
about 2.37 and 5.28 Mg/ha for currently available and future technology,
respectively (Table III). The offset of carbon is about twice as high for
electricity from SRWC as it is for ethanol from SRWC.
U.S. Potential Carbon and Energy Benefit
The potential carbon mitigation and energy supply benefits of SRWC are a
function of (1) the land base dedicated to the production of SRWC crops and
(2) the advancement of SRWC and associated energy-conversion technologies. To
evaluate the U.S. potential, two SRWC adoption scenarios are considered.
The first is a "current conditions" scenario. Table IV gives the
assumptions on land base and yield used in this scenario. The land base
selected (14 x 106 ha) is the amount of land currently eligible for CRP in
SRWC-suitable regions. As noted previously, SRWC is potentially profitable
today on this land base if government CRP subsidies are included (Lothner et
al. , 1988) . The selected yield is the same as previously stated; the
projected U.S. average of all lands capable of SRWC yields in excess of
11 Mg/ha/year, The previously described current energy-conversion
technologies are also assumed. Under these assumptions, the annual carbon
5-99
-------
18
offset by SRWC-based energy production is 33.2 x 106 Mg C/year if the wood is
used to produce ethanol and 73.1 x 106 Mg C/year if the wood is used for
electricity.
The second scenario is a "future" scenario. The assumptions include a
doubling of the SRWC-dedicated land base to 28 x 106 ha and targeted
improvements in SRWC and associated conversion technologies (Table IV). Under
these assumptions, the total U.S. carbon benefit (offset) would be 148 x
106 Mg/year with ethanol production, or 242 x 106 Mg/year with combustion and
electricity generation (Table IV). This is about 13 or 20% of the current
U.S. fossil fuel carbon emissions (ethanol or electricity, respectively) and
2.9 or 4.4% of global fossil fuel carbon emissions.
Discussion
The results of our analysis suggest that SRWC have significant potential
to offset annual U.S. fossil fuel carbon emissions. But these results oust be
considered in light of our assumptions. In both scenarios we assume a
considerable land base in the same order of magnitude as is currently
dedicated to corn or soybeans. A land use change of this magnitude would have
profound effects on local economies and perhaps on national agricultural
markets. The important point is that large amounts of agricultural quality
land are needed if SRWC technology is to significantly contribute to U.S.
energy needs and carbon dioxide mitigation. Such shifts in land use have
occurred in the recent past. In the early 1950s soybeans were an
5-100
-------
insignificant part of the U.S. agricultural sector; in 1988 soybeans were
planted on 24 million hectares.
We also assume that average yields of 14 to 22 Mg/ha/year can be
obtained on large acreages in the United States without irrigation. This
would appear plausible given the research experience under current climatic
conditions, but if the Midwest becomes drier as a result of greenhouse
warning, the acreage of land that could support reasonable SRWC yields without
irrigation would dramatically decrease.
Although this paper addresses the use of SRWC to provide biomass
feedstock, we would be remiss if we did not acknowledge the importance of such
other sources of biomass as crop and forestry residues and herbaceous energy
crops. Forestry residues are widely used throughout the forest industry to
supply power to pulp and paper facilities and mills. Existing forests are
being harvested to support power plants in northern New England. Because of
the much lower yields in most natural forests (typically 2-4 Mg/ha/year), the
use of traditional natural forests to supply wood feedstock on a sustainable
basis would require a much larger land base than would SRWC. The
environmental and societal impacts of using existing forests could also be
considerable, although with careful management the New England experience
appears to have been fairly positive (Hudson and Mullett, 1987).
Herbaceous energy crops have advantages and disadvantages relative to
SRWC. Herbaceous energy crops can probably be used over a wider range of
sites because some of the crops can be grown on drier and steeper settings
than SRWC. Herbaceous energy crops also are more similar to conventional
agricultural crops and thus easier for farmers to integrate into their
5-101
-------
20
existing operations. Herbaceous energy crops will also provide an economic
return to the farmer much more quickly than SRWC.
Disadvantages of herbaceous energy crops include higher erosion
potential (if the energy crop is sorghum or corn), higher inputs of
fertilizers, greater harvest/transportation/storage losses (if handled as a
hay crop), and lower energy content per unit of dry matter [17.4 GJ/Mg vs
19.8 GJ/Mg (Turhollow and Perlack, 1991)].
The carbon offset per unit of land dedicated to herbaceous crops will be
less than if that land was dedicated to SRWC unless the herbaceous yields are
higher. Using the production scenarios and fossil carbon inputs for SRWC,
sorghum, and switchgrass presented in Turhollow and Perlack (1992), assuming
identical gross yields (14 dry Mg/ha), and burning the feedstock for
electricity at equal conversion efficiencies, we calculate that on a per
hectare basis, switchgrass could displace 3.81 Mg C/ha/year; sweet sorghum,
4.65 Mg/ha/year; and SRWC, 5.22 Mg C/ha/year (Table V) . Undoubtedly, in some
(and perhaps most) situations herbaceous energy crops would be preferable to
SRWC but the preference would be strongly affected by potential yield, the
existence of an infrastructure for handling SRWC, and the motivations for
growing the crops.
The per hectare carbon savings of a corn-to-ethanol energy strategy are
relatively low compared with a SRWC-to-ethanol energy strategy. This is
because the fossil energy inputs required to grow corn and convert it to
ethanol are quite high. Using information from Marland and Turhollow (1991)
on carbon inputs in corn production and conversion to ethanol and the
substitution properties of ethanol for gasoline, we calculate that an average
5-102
-------
hectare yielding 7.47 Mg corn/ha/year would produce 2,783 L of ethanol and
give a net carbon displacement of 0.54 Mg C/ha/year if most of the processing
energy was coal-generated electricity (as is the case now) and 1.3 Mg
C/ha/year if all the processing energy was derived from bioraass (Table V).
Both these estimates include the backing out some of the energy inputs to
processing in recognition of the byproducts of corn-to-ethanol production
(analogous to our electricity credits in the ethanol scenarios) and the
substitution of ethanol for gasoline. In comparison with a hectare of SRWC, a
hectare of corn would produce 71% of the ethanol and 23% of the carbon
savings, assuming current technology for both SRWC and corn. From an
environmental perspective, corn production for ethanol is also probably less
desirable compared with either SRWC or switchgrass because of the greater
input of fertilizer and pesticides and the greater erosion potential,
In summary, our analysis suggests that the use 6f SRWC to produce
electricity or ethanol could make an important contribution to the U.S. energy
sector and to the reduction of U.S. fossil fuel carbon emissions. However,
that contribution depends on the dedication of a significant fraction of the
U.S. agricultural land base to energy crops and continued improvements in SRWC
production technology and biomass-energy- conversion technologies.
5-103
-------
Acknowledgments
The careful reviews by Greg Marland and Bob Cushman in the Carbon
Dioxide Program at Oak Ridge National Laboratory contributed significantly to
the accuracy of the information in this paper. We also acknowledge the
thoughtful suggestions of two anonymous reviewers. Parts of this paper were
presented at the North American Conference on Forestry Responses to Climate
Change, Washington, D.C., May 15-17, 1990, sponsored primarily by the Climate
Institute.
References
Combustion Engineering: 1981, 'Chapter 1: Steam Power Plant Design' in:
Combustion Fossil Power Systems. Combustion Engineering/Rand McNally.
Windsor, CT.
DOE/BMWTD: 1988, U.S. Department of Energy, Biofuels and Municipal Waste
Technology Division, DOE/CH10093-25. 'Five-year Research Plan
1988-1992' .
Hinman, N. D., Schell, D. J., Riley, C. J., Bergeron, P. W., and Walet, P. J.
1991, 'Preliminary Estimate of the Cost of Ethanol Production for SSF
Technology' in R. H. Texiera and B. J. Goodman (eds.), Ethanol Annual
Report FY 1990. SERI/TP-260-3674. Solar Energy Research Institute,
Golden, CO.
5-104
-------
Hudson, N., and Mullett, D.: 1987, 'Wood-Fired Electric Plants and Sound
Silvicultural Practices: A Working Formula for Happy Coexistence, as
Adopted and Successfully Implemented in the State of Vermont', internal
report, Vermont Department of Forests, Parks and Recreation, Montpelier.
Larson, E. D., and Svenningsson, P.: 1990, 'Development of Biomass
Gasification Systems for Gas Turbine Power Generation', paper presented
at Energy from Biomass Wastes XIV, January 29-February 2, 1990, Lake
Buena Vista, FL.
Larson, E. D., and Williams, R. H. : 1990, ' Biomass.-Fired Steam-Injected Gas
Turbine Cogeneration', Biologue 6(5), 12-19.
Lothner, D. C., Hansen, E. E., and Netzer, D. A.: 1988, 'Growing and
Utilizing Intensively Cultured Woody Crops for Energy: Some Recent
Evidence from the North Central United States' in G. Lonner and
A. Tornquist (eds.), Proc. IEA Bioenerpv. Task III. Activity A. Workshop
Economic Evaluations of Biomass Oriented Systems for Fuel. Swedish
University of Agricultural Sciences, Uppsala, Sweden, pp. 92-102.
Lynd, L. L., Cushman, J. H., Nichols, R. J., and Wyman, C. E.: 1991, 'Fuel
Ethanol from Cellulosic Biomass', Science 251, 1318-1323.
Marland, G.: 1983, 'Carbon Dioxide Emission Rates for Conventional and
Synthetic Fuels', Energy 8(12), 981-992.
Marland, G., and Turhollow, A. F.: 1991, 'C02 Emissions from the Production
and Combustion of Fuel Ethanol from Corn', Energy 16, 1307-1316.
5-105
-------
New Farm and Forest Task Force: 1987, 'New Farm arid Forest Products, Responses
to the Challenges and Opportunities Facing American Agriculture',
report to the Secretary, U.S. Department of Agriculture. U.S.
Government Printing office 1987 - 721-254-1302/60251, Washington, DC.
Ostlie, L. D.: 1989, 'The Whole Tree Burner: A New Technology in Power
Generation', Biologue 5(3), 7-9.
Ranney, J. W., Barkley, B. A., Turhollow, A. F., and Granger, C.: 1987, 'Short
Rotation Intensive Culture Economics for Energy in North America', Proc.
i
Int. Energy Agency Workshop on Economic Evaluations of Short Rotation
Biomass Systems. August 11-13, 1987, Duluth, MN. USDA Forest Service,
North Central Forest Experiment Station, Duluth.
Stokes, B. J., Frederick, D. J., and Curtin, D. T.: 1986, 'Field Trials of a
Short-Rotation Biomass Feller Buncher and Selected Harvesting Systems',
Biomass 11, 185-204.
Turhollow, A. F., and Perlack, R.D.: 1991, 'Emissions of C02 from Energy Crop
Production', Biomass and Bioener SX 1, 129-135.
USDA/ERS (U.S. Dept. of Agric., Economic Resources Service): 1989,
'Agricultural Resources: Cropland, Water and Conservation, Situation
and Outlook Report', AR-16, September 1989, USDA Economic Research
Service, Resource and Technology Division, Washington, DC.
USDA/SCS (U.S. Dept. of Agric., Soil Conservation Service): 1987, 'Basic
Statistics, 1982 National Resource Inventory', Statistical Bulletin
Number 756, USDA Soil Conservation Service, Resource Inventory Division,
Washington, DC.
5-106
-------
USDA/SCS (U.S. Dept. of Agric., Soil Conservation Service): 1989, 'Summary
Report, 1987 National Resource Inventory', Statistical Bulletin 790,
Iowa State Statistical Laboratory. USDA Soil Conservation Service,
Resource Inventory Division, Washington, DC.
Wright, L. L. , Doyle, T. W., Layton, P. A., and Ranney, J. W.: 1989, 'Short
Rotation Woody Crops Program: Annual Progress Report for 1988',
ORNL-6594. Oak Ridge National Laboratory, Oak Ridge, TN.
Wright, L. L., and Ehrenshaft, A.: 1990, 'Short Rotation Woody Crops Program:
Annual Progress Report for 1989', ORNL-6625, Oak Ridge National
Laboratory, Oak Ridge, TN.
Wright, L. L., Graham, R. L., Turhollow, A. F., and English, B. C.: 1992,
'Opportunities to Mitigate Carbon Dioxide Buildup Using Short-Rotation
Woody Crops', in R. N. Sampson and D. Hair (eds.), Forests and Global
Warming, Vol. 1. American Forestry Association, Washington, DC.,
pp. 123-156,
5-107
-------
26
TABLE I: Short-rotation woody crop yields by U.S. region - current and future
expected yields for operational conditions and maximum observed values (Wright
et al., 1992)
Yields
(dry) Mg/ha/year
Region Current Goal Maximum observed
Northeast
9.0
15.7
15.7
South/Southeast
9.0
17.9
15.7
Midwest/Lake
11.2
20.2
15.7
Northwest
15.7
29.1
43.3
Subtropics
15.7
29.1
27.6
5-108
-------
27
TABLE II: Annual carbon offset per hectare by electricity produced from short
rotation woody crops (SRWC)
Assumptions and results Technology status
Available Future
Biomass yield (before losses), (dry) Mg/ha/year 14.0 22.4
Bioraass yield (after losses), (dry) Mg/ha/year 11.3 18.5
Conversion efficiencies assumed
wood-to-electricity (%) 33 42
coal-to-electricity (%) 33 42
Electricity production, MWh/ha/year* 20.5 42.7
Gross carbon offset by fuel substitution, 5.51 9.03
(Mg/ha/year)**
SRWC production carbon emissions, Mg/ha/year 0.29 0.40
Net carbon offset, Mg/ha/year 5.22 8.63
Assumes wood energy content of 19.8 GJ/dry Mg
Assumes coal carbon content of 24.65 kg C/GJ.
5-109
...fc.
-------
28
TABLE III: Annual carbon offset per hectare by short rotation woody crops
(SRWC) ethanol production
Assumptions and results
Technology
status
Current
Future
Biomass yield (before losses), (dry) Mg/ha/year
14.0
22.4
Biomass yield (after losses,) (dry) Mg/ha/year
11.3
18.5
Wood-to-ethanol conversion efficiency (%)
41
60
Ethanol yield: L/(dry) Mg biomass
L/ha
344
3889
503
9312
Electricity yield : kWh/(dry) Mg biomass
: kWh/ha
184
2079
101
1868
Carbon offset by fuel substitution, Mg/ha/year
ethanol substitution*
electricity substitution*1
2.25
0.41
5.39
0.29
SRWC production carbon emissions, Mg/ha/year
0.29
0.40
Net carbon offset, Mg/ha/year
2.37
5.28
Assumes the carbon content of gasoline is 0.723 kg/L and one liter of
ethanol can substitute for 0.8 L of gasoline.
# Assumes 1.0 kWh of biomass-derived electricity substitutes for 0.75 kWh of
coal-derived electricity, a coal conversion efficiency of 33% (current) and
42% (future), and a carbon content of coal of 24.65 kg/GJ.
5-110
-------
TABLE IV: U.S. production scenarios — land and yield assumptions: carbon
offset, ethanol, and electricity generation; comparison to U.S. fuel and
electricity consumption; and comparison to U.S. and global carbon emissions
from fossil fuel
Short rotation woody crop energy scenarios
Current Current Future Future
ethanol electricity ethanol electricity
Net Yield, (dry) Mg/ha/year 11.3
U.S. land base, 106 ha 14
11.3
14
18.5
28
18.5
28
Ethanol generated, 109 L/yr 54.3
Electricity generated 29.1
(106 HWh)
Percentage of current 10.4
U.S. gasoline consumption
Percentage of U.S. electric 1.1
power consumption
287.1
10.7
260.7
52.4
49.7
2.0
1196.6
44.5
Carbon offset, 10B Mg/year 33.2
Percentage of U.S. fossil 2.9
fuel carbon emissions
Percentage of world fossil 0.6
fuel carbon emission's
73.1
6.0
1.3
148
13.2
2.9
242
19.9
4.4
5-111
-------
Table V: A comparison of short rotation woody crops (SRWC) derived energy with other forms of biomass
energy. Combustion of SRWC is compared with combustion of sweet sorghum or switchgrass. Differences in net
yield are a function of differences in harvesting and transportation losses. Differences in fossil carbon
inputs are a function of differences in crop management (e.g., fertilizer use, tillage, pesticides).
Conversion of SRWC to ethanol using current technology is compared with conversion of corn grain to ethanol
using current technology and assuming current use of coal for process energy or assuming use of biomass for
process energy
Gross yield
Net yield
Electricity
or ethanol
production
Gross fossil
carbon
displacement
Fossil carbon
inputs
Net carbon
savings
Combustion
Mg/ha/year
Mg/ha/year
Mwh/ha/year*
Mg C/ha/year
Mg C/ha/year
Mg C/ha/year
SRWC
14
11.3
20.5
5.51
.29
5.22
Sweet sorghum
14
11.8
18.8
5.06
.41
4.65
Switchgrass
14
9.64
15.4
4.13
.32
3.81
Ethanol
production
Mg/ha/year
Mg/ha/year
L EtOh/ha/year
Mg C/ha/year
Mg C/ha/year
Mg C/ha/year
SRWC
14
11.3
3882
2.66
0.29
2.37
Corn grain**
-
7.47
2783
1.61
1.07
0.54
Corn grain***
-
7.47
2783
1.61
0.31
1.30
Assumes a conversion efficiency of 33%.
Assuming 75% of electricity needed for process energy is derived from coal combustion.
*** Assuming all electricity needed for process energy is derived from combustion of biomass.
-------
SESSION VI: ENERGY SOURCES/SOLAR/RENEWABLE
Robert Williams, Chairperson
ROLES FOR BIOMASS ENERGY
IN SUSTAINABLE DEVELOPMENT
Robert H. Williams
Center for Energy and Environmental Studies
Princeton University
Princeton, NJ 08544
INTRODUCTION
Biomass (plant matter) has been used as fuel for millenia. In the 18th and 19th centuries it was widely
used in households, industry, and transportation. In the US, as late as 1854, charcoal still accounted for nearly
half of pig iron production, and throughout the antebellum period wood was the dominant fuel for both
steamboats and railroads [1], Biomass dominated global energy consumption through the middle of the 19th
century [2]. Since then biomass has accounted for a diminishing share of world energy, as coal and later oil and
natural gas accounted for most of the growth in global energy demand. Today biomass is not much used by
industry, though it is still widely used for domestic applications in developing countries-especially in rural areas
[3]. Still, biomass accounts for about 15% of global energy use, only slightly less than the share of global
energy accounted for by natural gas (see Figure /).
Although the trend has been away from biomass as an energy source, there are strong reasons for
revisiting biomass energy:
0 Dependence on liquid hydrocarbon transport fuels has led to urban air pollution problems in many areas
that cannot be solved simply by mandating further marginal reductions in tailpipe emissions. California
has adopted a policy mandating the phased introduction of low- and zero-emission transport vehicle/fuel
systems, and many other US states and some other countries are likely to pursue similar policies [4],
Some biomass-based transport energy options could effectively address this challenge r5,6].
o The prospect of declining future production of conventional oil in most regions outside the Middle East
[7] once more raises concerns about the security of oil supplies. Fluid fuels derived from biomass
substituted for imported oil can help reduce energy security risks [5].
o Responding to concerns about global warming may require sharp reductions in the use of fossil fuels
[8]. Biomass grown sustainably and used as a fossil fuel substitute will lead to no net buildup in
atmospheric carbon dioxide, because the C02 released in combustion is compensated for by the C02
extracted from the atmosphere during photosynthesis.
o A major challenge facing developing countries is to find ways to promote rural industrialization and
rural employment generation, to help curb unsustainable urban migration [9]. Low-cost energy derived
from biomass sources could support such activities [5].
o There are large amounts of deforested and otherwise degraded lands in tropical and subtropical regions
in need of restoration [10]. Some of these lands could be restored by establishing biomass energy
plantations on them. Part of the revenues from the sale of biomass produced on such lands could be
used to help pay for these land restoration efforts [3,5].
o In industrialized countries, efforts to provide food price and farmer income stability in the face of
growing foodcrop productivities have led to a system of large-scale agricultural subsidies. Despite
mounting economic pressures to reduce or eliminate such subsidies, so doing is difficult politically [11].
However, converting excess agricultural lands to biomass production for energy would both provide a
6-1
-------
new livelihood for farmers and make it possible to phase out such subsidies [3,5].
Because of such considerations, and because of good prospects for providing competitive energy
supplies from biomass using modem energy conversion technologies, it was projected in a recent study exploring
the prospects for renewable energy that biomass can have major roles as a renewable energy source [5]. In a
renewables-intensive global energy scenario constructed for that study it was estimated that renewable energy
could provide about 45% of global primary energy requirements in 2025 and 57% in 2050, with biomass
accounting for about 65% of total renewable energy in both years (see Figure 2). For the US, the coireponding
renewable energy shares of total primary energy were projected to be similar to the renewable shares at the
global level, with biomass accounting for 55-60% of total renewable energy in this period (see Figure 3).
In the renewables-intensive global energy scenario, biomass supplies are provided mainly by biomass
residues of ongoing agricultural and forest product industry activities (e.g., sugar cane residues and mill and
logging residues of the pulp and paper industry) and by feedstocks grown on plantations dedicated to the
production of biomass for energy purposes. In the present analysis, the discussion is focussed on plantation
biomass, which accounts for about three-fifths of global biomass supplies in the period 2025-2050 [5].
THE CHALLENGES POSED BY BIOMASS ENERGY
The notion of shifting back to biomass for energy flies in the face of conventional wisdom. Bringing
about such a shift would require overcoming strong beliefs held be many people that biomass is inherently
unpromising as an energy supply source. It is widely believed that:
o Biomass is an inconvenient energy carrier and thus inherently unattractive for modern energy systems,
o The use of land to grow biomass for energy conflicts with land needs for food production,
o Large-scale use of biomass for energy would create environmental disasters,
o The energy balances associated with biomass production for energy are unfavorable,
o Biomass energy is inherently more costly than fossil fuel energy,
o Resource constraints will limit biomass to a minor role in a modem global energy system.
In what follows each of these concerns is dealt with in turn.
ATTRACTING CONSUMER INTEREST BY MODERNIZING BIOMASS ENERGY
Biomass is often called "the poor man's oil" [9]. This characterization arises in part from the low bulk
density of biomass fuels. Freshly cut wood typically has an energy density of about 10 GJ per tonne-compared
to 25 to 30 GJ per tonne for various coals and more than 40 GJ per tonne far oil; it is thus both difficult and
costly to transport biomass fuels long distances; in rural areas of developing countries women and children spend
considerable time gathering fuelwood for cooking. Wood cookstoves also pollute-generating in rural kitchens of
developing countries total suspended particulates, benzo-a-pyrcnes, and other pollutants-often at levels far in
excess of ambient air quality standards [12],
Also, as incomes rise, consumers' preferences shift to energy carriers of higher quality. The higher the
quality of the fuel, the more convenient it is to use and the less the pollution generated-hence the more desirable
it is. This phenomenon is well-known for cooking fuels: charcoal is preferred to wood; kerosene is preferred to
charcoal; and clean gaseous fuels such as liquid petroleum gas (LPG) are preferred to kerosene [12].
It is often argued that the existence of this "energy ladder" shows thai consumers will shift away from
-------
biomass fuels, as they are able to afford energy carriers of higher quality. Data on the global patterns of
biomass use for energy support this contention: while biomass accounts for 38% of energy use in developing
countries, where it is used mainly by poor people in rural areas, it accounts for less than 3% of energy use in the
industrialized countries (see Figure 1).
However, this argument concerns not biomass per se but rather the energy carriers used by the
consumer. Biomass can be used directly as a low-quality solid fuel (e.g. fuelwood), it can be upgraded into a
higher quality solid fuel (e.g. charcoal), or it can be converted into gaseous or liquid fuels or electricity. In
short, biomass can be utilized as an energy source with a wide range of energy carriers. Thus biomass would be
an acceptable energy source at higher income levels if it could be converted into energy carriers deemed
desirable by consumers with higher incomes.
addressing the food vs. fuel controversy
The renewables-in tensive global energy scenario developed in [5] calls for establishing worldwide some
400 million hectares of biomass plantations for energy by the second quarter of the 21st century—a land area that
is not small compared to the nearly 1500 million hectares now in cropland [13]. Because the world population is
expected to nearly double by that time, the potential for conflict between biomass production for food and
biomass production for energy warrants careful scrutiny. Because land is needed to grow food, but energy can
be provided in many ways, food production should have priority. The key questions are: How much land is
needed for food production? And how does this need compare with the arable land resource? In addressing
these questions it is useful to consider the industrialized and developing country situations separately.
Industrialized Countries
Because their population growth is slow and food yields have been increasing, the amount of land
needed for food production is declining in industrialized countries.
In the US, more than one-fifth of total cropland, some 33 million hectares, was idled in 1990, either to
support crop prices or to control erosion. Furthermore, the Soil Conservation Service of the U.S. Department of
Agriculture projects that the amount of idle cropland will probably increase to 52 million hectares by 2030 as a
result of rising crop yields, despite an expected doubling of exports of maize (com), wheat, and soybeans in this
period [14], The urgency of addressing the challenge of excess agricultural lands was the major theme of the
1987 report of The New Farm and Forest Products Task Force to the Secretary of Agriculture [15]:
"The productive capacity of US agriculture is greatly underutilized. The country today has carryover
stocks of between six months and one year's production of major commodities, with productivity
continuing to increase at a faster rate than demand. Estimates of land in excess of production needs to
meet both domestic and export market demand range as high as 150 million acres [61 million
hectares]-with about one third of that already available from the Conservation Reserve Program. This
represents an enormously wasted national asset which, if transformed into a more productive one
through new products, would have a profoundly positive impact on the Nation's economy."
In the European Community more than 15 million hectares of land will have to be taken out of farming
by 2000 if surpluses and subsidies associated with the Common Agricultural Policy are to be brought under
control [16]; a study carried out by the Netherlands Scientific Council for Government Policy projects that the
land needed for food projection in the Community could be 50 to 100 million hectares less in 2015 than at
present [17].
While the conversion of excess cropland in the industrialized countries to energy plantations presents an
opportunity to make productive use of these lands, such a conversion cannot be easily accomplished under
present policies. In many countries farmers are deterred by a subsidy system that specifies what crops the
farmer can produce in order to qualify for a subsidy; and energy crops are not allowed. Conversion to profitable
6-3
-------
biomass energy production will make it possible eventually to phase out many of these agricultural subsidies (see
Appendix A). However, this will not be accomplished overnight, because of the economic dislocations that
would result; today these subsidies total about 5300 billion per year for countries of the Organization for
Economic Cooperation and Development (OECD) [11]. As long as a system of subsidies continues, however,
the bias against energy crops should be removed.
Developing Countries
For developing countries the situation is quite different. Because of expected population growth and
rising incomes, it is likely that more land will be needed for food production. The Response Strategies Working
Group of the Intergovernmental Panel on Climate Change has projected that the land in food production in
developing countries will increase 50% by 2025 from the present level of about 700 million hectares (see Table
1) (18]. The demand can be compared to potential supply-that is, land physically capable of supporting
economic crop production, within soil and water constraints. Potential cropland was estimated for 91 developing
countries in a 1991 Food and Agriculture Organization (FAO) study [19]. For these countries potential cropland
was estimated to be about 2055 million hectares-nearly three times present cropland (see Table 1).
Looking to the year 2025 and assuming cropland requirements in developing countries increase 50% by
then, there would still be a substantial surplus potential cropland of nearly 1,000 million hectares in these
countries (see Table 1). There would be substantial regional differences, however, with major surpluses totalling
more than 1,100 million hectares in Latin America and Africa, and a 110 million hectare deficit in Asia. (China
was not included in the FAO analysis.) Thus it appears that substantial amounts of land suitable for energy
plantations may be available in both Latin America and sub-Saharan Africa, even with major expansions of
cropland to feed the growing population. But in Asia, with its high population density, conflicts with food
production could become significant.
The extent of the potential conflict in Asia, however, depends sensitively on future food crop
productivities. It is possible that apparent food/fuel conflicts could instead prove to be synergistic, if some of the
energy produced in biomass plantations woe used to help make agriculture more productive. Increasing food
production on the better lands, while growing trees or perennial grasses for energy on marginal lands, would
generally be environmentally preferable to increasing agricultural output by bringing marginal lands into food
production. Detailed assessments are needed, on a country-by-country basis, to understand the prospects for
productivity gains with more intensive agricultural management and the extent of food/fuel conflict if the
agricultural sector is more intensively managed.
Unfortunately, the FAO study does not clarity where new cropland would come from. To be sure, some
forestlands are involved. Clearly, it would not be desirable to cut down virgin forests in flavor of intensively
managed biomass plantations. Cutting down virgin forests could be avoided, however, by targeaing for biomass
plantations lands that are deforested or otherwise degraded and that are suitable for reforestation. Large land
areas have been degraded. One estimate is that 2,077 million hectares of tropical lands are degraded, of which
758 million hectares are judged suitable for reforestation (see Table 2).
Outside of Asia the amount of degraded land suitable fa- reforestation (excluding degraded lands in the
desertified drylands category, one-fifth of which is estimated to be suitable for reforestation--.ree Table 2) is
substantial (see Table 2)~some 156 million hectares in Latin America (49% of the total degraded land area in
Latin America) and 101 million hectares in Africa (12% of the total degraded land area in Africa). In Asia, such
land areas are also large-some 169 million hectares (18% of the total degraded land area); however, for Asia,
country by country assessments are needed to determine the extent to which its degraded lands will be needed
for food production or other purposes warranting higher priority than energy. China, despite its high population
density, has a goal of increasing forest cover by 52 million hectares by 2000 (compared to the mid-1980s) and
by an additional 93 million hectares over the longer term [20].
The main technical challenge of restoration is to find a sequence of plantings that can restore ground
temperatures, organic and nutrient content, moisture levels, and other soil conditions to a point where crop yields
6-4
-------
are high and sustainable. Successful restoration strategics typically begin by establishing a hardy species with
the aid of commercial fertilizers or local compost. Once erosion is stabilized and ground temperatures lowered,
organic material can accumulate, microbes can return, and moisture Mid nutrient properties can be steadily
improved. This can lead to a self-regenerating cycle of increasing soil fertility [21,22].
If it is feasible to overcome this technical challenge and various other socioeconomic, political and
cultural challenges [3], plantation biomass in developing regions could make substantial contributions to world
energy without serious conflict with food production. In sub-Saharan Africa and Latin America, where potential
land areas for plantations are especially large, biomass could be produced by the second quarter of the 21st
century in quantities large enough to make these regions major exporters of biomass-derived liquid fuels (see
Figure 4), offering competition to oil exporters and bringing price stability to the global liquid fuels; market [5].
Converting such large areas of degraded lands to successful commercial plantations would be a
formidable task. Research is needed to identify the most promising restoration techniques for all the different
land types and conditions involved. Yet the fact that many of the successful plantations in developing countries
have been established on degraded lands [3] suggests that it may be feasible to deal with these challenges, with
adequate research and commitment. And interest in restoring tropical degraded lands is high, as indicated by the
ambitious global net afforestation goal of 12 million hectares per year by the year 2000 set forth in the
Noordwijk Declaration at the 1989 Ministerial Conference on Atmospheric and Climate Change in Noordwijk,
The Netherlands [23].
MAKING BIOMASS PRODUCTION FOR ENERGY ENVIRONMENTALLY ATTRACTIVE
Throughout the 19th and 20th centuries, there has been substantial deforestation worldwide, as a result
of both land clearing for agriculture and the non-sustainable mining of the forests for forest products. The more
simplified landscape resulting from this deforestation is unable to support the diversity of species that once
flourished there. Moreover, modern intensive agricultural management practices have created other serious
environmental problems-including the loss of soil quality as a result of continual mining of the soil with plant
harvesting, erosion as a result of intensive cultivation practices, and contamination of runoff with nitrates and
other chemicals arising from the use of fertilizers, herbicides, and pesticides. A major concern is that such
problems would be aggravated by a major shift to biomass energy.
There is no doubt that biomass can be grown for energy purposes in ways that are environmentally
undesirable. However, it is also possible to improve the land environmentally through the production of biomass
for energy. The environmental outcome depends sensitively on how the biomass is produced.
Consider first the challenge of sustaining the productivity of the land. Since the harvesting of biomass
removes nutrients from the site, care must be taken to ensure that these nutrients are restored. In various ways
this challenge can be dealt with for energy plantations mar easily than is feasible in agriculture or in industrial
fiber production, largely as a result of the fact that energy markets allow flexibility in the choice of biomass
feedstocks, so that choices can be made to better meet environmental objectives. This is especially true for
biomass conversion technologies that begin with thermochemical gasification (which will often be die preferred
approach for providing modem energy carriers from biomass feedstocks [5]); such processes can accommodate a
wide range of alternative feedstocks.
With thermochemical gasification ii is feasible to recover all mineral nutrients as ash from the gasifier at
the biomass conversion facility and to return the ash to the plantatation site for use as fertilizer. Of course, fixed
nitrogen lost to the atmosphere at the conversion facility must be replenished. However, there are several ways
this can be accomplished in environmentally acceptable ways. First, when trees are the harvested crop, the
leaves, twigs, and small branches, in which nutrients are concentrated, can be left at the site to reduce nutrient
loss. (So doing also helps maintain soil quality and reduce erosion through the addition of organic matter to the
soil.) Also, nitrogen-fixing species can be selected for the plantation or for interplanting with the primary
plantation species to eliminate or reduce to low levels the need for artificial fertilizer inputs. The promise of
6-5
4.
-------
intercropping strategies is suggested by 10-year trials in Hawaii, where yields of 25 dry tonnes per hectare per
year were achieved without nitrogen fertilizer when Eucalyptus was interplanted with nitrogen-fixing Albiiia
trees [24], Biomass production for energy allows much more flexibility than is possible with agriculture in
meeting fixed nitrogen requirements this way. In agriculture, the market dictates the choice of feedstocks with a
narrow range of acceptable characteristics, but the conversion technology usually puts few restrictions on the
choice of biomass feedstock for energy systems, aside from the requirement of high productivity, which is
needed to keep costs at acceptable levels.
Energy crops also offer flexibility in dealing with erosion and with chemical pollution from herbicide
use. These problems occur mainly at the time of crop establishment. Accordingly, if the energy crop is an
annual crop (e.g., sweet sorghum), the erosion and herbicide pollution problems would be similar to those for
annual row-crop agriculture. The cultivation of such crops should be avoided on erodibie lands. However, the
choices for biomass energy crops also include fast-growing trees that are harvested only every 5 to 8 years and
replanted perhaps every 15 to 24 years and perennial grasses that are harvested annually but replanted perhaps
only once in a decade. In both cases erosion would be sharply reduced, on average, as would the need for
herbicides.
A major concern about agriculture is water pollution from nitrate runoff associated with the excessive
use of chemical fertilizers. Where it would not be practical to deal with this problem by planting nitrogen-fixing
plantation species as an alternative to chemical fertilizers, runoff pollution could be controlled instead by planting
fast-growing trees having low nitrogen-use efficiency in riparian zones [25]. In the future it will also be possible
to use "designer" fertilizers whose release is timed to match the temporal variations in the plant's demand for
fertilizer [26,27].
Another concern is chemical pollution from the use of pesticides to control the plantation crop against
attack by pests and pathogens. While plantations in the tropics and subtropics tend to be more affected by
disease and pest epidemics than those in temperate regions, experience with plantations in these regions shows
that careful selection of species and good plantation design and management can be helpful in controlling pests
and diseases, rendering the use of chemical pesticides unnecessary in all but extraordinary circumstances. A
good plantation design, for example, will include: (i) areas set aside for native flora and fauna to harbor natural
predators for plantation pest control, and perhaps (ii) blocks of crops characterized by different clones and/or
species. If a pest attack breaks out on one block, a now common practice in well-managed plantations is to let
the attack run its course and to let predators from the set-aside areas help halt the pest outbreak [3].
Biomass plantations are often criticized because the range of biological species they support is much
narrower than for natural forests. While this is generally true, the criticism is not always relevant It would be
relevant if a virgin forest were replaced with a biomass plantation. However, it would not be relevant if a
plantation and associated natural reserves were established on degraded lands; in this instance, the restored lands
would be able to support a much more diverse ecology than was possible before restoration. If biomass energy
crops were to replace monocultural food crops, the effect on the local ecology would depend on the plantation
crop species chosen, but in many cases the shift would be to a less ecologically simplified landscape [28].
As already noted, establishing and maintaining natural reserves at plantations can be helpful in
controlling crop pests while providing local ecological benefits. However, preserving biodiversity on a regional
basis will require land-use planning in which natural forest patches are connected via a network of undisturbed
corridors (riparian buffer zones, shelterbelts, and hedgerows between fields), thus enabling species to migrate
from one habitat to another [28].
While major expansions in research are needed to provide a sound analytical and empirical basis for
achieving and sustaining high biomass yields in environmentally acceptable ways, there is lime for this research
and extensive field trials, because major bioenergy industries can be launched using as feedstocks primarily
residues from the agricultural and forest products industries [5]. If substantial commitments are made to biomass
plantation research in the near term, plantation biomass could start to make contributions to energy when residue
6-6
-------
supplies are no longer adequate to meet the needs of the growing biomass energy industry, near the turn of the
century or shortly thereafter [3].
ACHIEVING FAVORABLE ENERGY BALANCES IN BIOMASS PRODUCTION FOR ENERGY
For biomass energy systems to be viable, the net energy balance must be favorable-i.e., the useful
energy produced must be greater than the fossil fuel energy inputs required to provide the biomass energy.
Concerns about net energy balances have been widely voiced in the case of fuel ethanol from maize (corn),
which is produced in the US under subsidy at a rate of 3 billion liters per year. In this case the net energy
balance is often marginal, and in some instances, the fossil fuel inputs to the system are greater than the alcohol
energy produced [29]. Maize, however, is a feedstock intended primarily for use as food, not fuel, and its
production system is especially energy-intensive among biomass crops.
There are many alternative biomass energy systems characterized by favorable net energy balances. For
example, the production of fuel ethanol from sugar cane in Brazil (where production at a level of 12 billion liters
per year provides nearly one-fifth of total transport fuel requirements) is characterized by an energy balance in
which the energy content of the produced alcohol averages about 6 times the fossil fuel inputs required to grow,
harvest, and transport the cane and convert it to alcohol [30]. Moreover, the net energy balances are also
favorable for many energy plantation crops that might be grown in temperate climates. Table 3 shows that with
present plantation technology, the energy content of the harvestable biomass would be in the range 11 to 16
times the fossil fuel energy needed to provide the biomass [31], and thai this ratio is expected to increase with
advanced technology. This harvested biomass could be used with near-term technology to produce methanol via
thermochemical gasification at an overall efficiency of about 63% [32], so that overall the amount of energy that
can be produced in the form of methanol is 7 to 10 times the fossil fuel input to the biomass energy production
system.
In general, energy systems having good prospects for becoming economically competitive tend to be
characterized by favorable net energy balances, while systems with poor economic prospects may have
unfavorable energy balances.
ACHIEVING ATTRACTIVE ECONOMICS BY MODERNIZING BIOMASS ENERGY
The planting, cultivation, and harvesting of biomass is generally more labor-intensive and costly than
recovering coal or other fossil fuels from the ground Thus, per unit of contained energy, biomass tends to be
the more costly, especially where there are abundant indigenous fossil fuel resources. This primary energy cost
comparison does not imply that biomass energy systems cannot be cost-competitive with fossil fuel energy
systems, however. A more meaningful measure of economic performance is the cost of the energy services
provided by the biomass and alternatives, taking into account the technologies for converting biomass into
modem energy carriers (electricity and gaseous or liquid fuels) and the energy end-use systems in which these
energy carriers would be used. On a cost-of-service basis the economic outlook for biomass can be favorable if
modern conversion and end-use technologies are used.
Biomass Electricity
Today biomass, mainly in the form of industrial and agricultural residues, is used to generate electricity
with conventional steam-turbine power-generators. These biomass power systems can be cost-competitive where
low-cost biomass fuels are available, in spite of the fact that steam-turbine technologies are comparatively
inefficient and capital-intensive at the small sizes required for biomass electricity production. The US currently
has more than 8,000 MW, of generating capacity fueled with such feedstocks, most of which was developed in
the 1980s. Electricity production based on this existing technology will not expand much in the future, however,
because unused supplies of low-cost biomass residues are rapidly becoming unavailable.
Biomass power generation involving the use of more costly but more abundant feedstocks could be
made cost-competitive by adapting to biomass advanced-gasification technologies originally developed for coal
6-7
-------
for use with gas turbine-based power systems [33]. Biomass is a more attractive feedstock for gasification than
coal because it is easier to gasify and has a very low sulfur content, so that expensive sulfur removal equipment
is not needed. Biomass integrated gasifier/gas turbine power systems with efficiencies of 40% or more will be
demonstrated in the mid-1990s and will probably be commercially available by 2000. These systems offer high
efficiencies and low unit capital costs for baseload power generation at relatively modest scales of 100 MW. or
less and will probably be able to compete with coal-fired power plants in many circumstances-even with
relatively costly biomass feedstocks. By 2025 conversion efficiencies as high as 57% may be feasible, using
advanced biomass gasification/ fuel-cell technologies, based on similar technologies (involving molten carbonate
and solid oxide fuel cells) being developed for coal [33].
The electric power industry is beginning to appreciate the importance of biomass for power generation.
In an assessment by the Electric Power Research Instititue of the potential for biomass-based power generation, it
is projected that biomass could be used to support 50,000 MW. of electric capacity in the US by 2010 and
probably twice that amount by 2030 [34].
Transport Fuels from Biomass
Unlike the auspicious near-term outlook for biomass-derived electricity, very large increases in the
world oil price are required before biomass-derived transport fuels could compete in cost with gasoline on a
cost-per-unit-of-fuel-energy basis. Nevertheless, because of ongoing changes in the transport sector, the
prospects are auspicious that biomass fuels will be able to compete in providing transport services at world oil
prices near the present low level. This prospect will be illustrated here for methanol and gaseous hydrogen fuels
derived from biomass via thermochemical gasification.
Based on the use of gasification technology1 that could be commercialized before the turn of the
century2 it should be feasible to produce either methanol or gaseous hydrogen delivered to the consumer at a cost
that is 40 to 50% higher than the price of gasoline near the turn of the century [6,32], when gasoline is expected
to have a pump price (exclusive of retail taxes) of $ 1.25 per gallon ($0.33 per liter). Even though these fuels are
likely to be more costly to produce than gasoline, the prospects are good that they would be able to compete on
a lifecycle cost basis (cents per km driven) with gasoline, if the biomass derived fuels were used, not in
internal-combustion-engine vehicles, but in fuel-cell vehicles.
Biomass-derived fuels are likely to be cost-compeiitive with the same fuels derived from coal even
when the biomass feedstock is more costly than coal [32], both because biomass contains much less sulfur (the
removal of which is costly) and because biomass is much more reactive and thus easier to gasify than coal. In
both cases the first step is thermochemical gasification aimed at producing a nitrogen-free synthesis gas (a
gaseous mixture consisting largely of carbon monoxide and hydrogen, with some methane). The relatively high
reactivity of biomass makes it possible to produce this synthesis gas with steam gasification at the relatively low
temperatures achievable in indirectly heated gasifiers (see footnote 1). However, because of coal's much lower
reactivity, much higher temperatures are required for efficient gasification-temperatures achievable only by
burning some of the coal in place with oxygen, in an "oxygen-blown" gasifier. The high cost of the required
oxygen plant more than offsets the lower cost of the coal feedstock.
While the internal combustion engine has dominated road transportation since the automobile was
introduced, the long-term outlook for this engine in transportation is clouded by growing concents about urban
1 The most promising technology involves an indirect gasifier, such 13 the indirectly heated gasifier being developed by the
Batielle Columbus Laboratory, in which biomass is gasified in Ream at relatively low temperatures (compared u> coal), using an external heat
source to provide the beat needed to drive (he endothennic reactions involved [29.32).
2 All technological component* required to produce either methanol or hydrogen other than the gasifier are already
well-established commercially.
6-8
-------
air quality. It is already becoming apparent in southern California that air quality goals cannot be met simply by
mandating further incremental reductions in tail-pipe emissions of new vehicles and that a shift to very-low- or
zero-emission vehicles is needed in order to reach these goals. Accordingly, the State of California has
mandated that 10% of new cars purchased in 2003 must be zero emission vehicles-a requirement that may be
adopted by many Eastern US states as well [4).
This California air-quality initiative has led to a substantial industrial effort to commercialize the
battery-powered electric car. While the battery-powered electric car is a zero-emission vehicle, this technology
will probably be limited to a modest fraction of the automotive fleet market in the long term, without major
advances in battery technology that make it feasible to overcome the long (several-hour) recharging time required
for batteries [5,6],
A zero-emission vehicle alternative to the battery-powered electric car is the fuel cell car operated on
compressed hydrogen. As in the case of the battery-powered electric car, electric motors provide the mechanical
power that drives the wheels. But in this case the electricity to run the motors is provided not by a battery but
rather by a fuel cell that converts energy stored in compressed hydrogen gas canisters directly into electricity.
Unlike the battery-powered electric car, the hydrogen fuel cell car need not be recharged but can be refueled in a
time comparable to that for a gasoline-fueled intemal-combustion-engine-powered car. Moreover, the lifecycle
cost of owning and operating a fuel cell car (in cents per km) operated on hydrogen derived from biomass would
probably be less than for a battery-powered electric car [S, 6, 32]. Although biomass-derived hydrogen is likely
to be atx>ut 50% more cosily than gasoline and although the hydrogen fuel cell car may cost 40% more than an
internal-combustion-engine car of comparable performance, on a lifecycle cost basis the hydrogen fuel cell
vehicle is likely to also be less costly-mainly because the fuel cell car is expected to be three times as
energy-efficient and because, like the battery-powered electric car, it is expected to have lower maintenance
costs; the energy conversion unit has fewer moving parts than in an internal-combustion-engine car and does not
have to be designed to contain explosions of fuel/air mixtures.
A drawback of the hydrogen fuel cell option is the requirement of a hydrogen gaseous fuel
infrastructure. This difficulty could be circumvented by using instead methanol as a hydrogen carrier the
methanol would be reacted with steam under the hood of the car and thereby converted into a mixture of carbon
dioxide and hydrogen-thereby providing the hydrogen needed to operate the fuel cell. There are three
advantages of using methanol instead of hydrogen: the fuel delivered to the consumer would be slightly cheaper
because, unlike hydrogen, the fuel does not have to be delivered pressurized; the car would be less expensive
because costly high-pressure hydrogen storage canisters would not be needed; and it is easier to establish an
infrastructure for marketing a liquid fuel like methanol than for a gaseous fuel like hydrogen. The drawbacks of
the methanol fuel cell option are; the energy conversion efficiency is less (the car is likely to be only 2.5 times
as energy-efficient as the gasoline internal-combustion-engine vehicle it would replace, owing to the energy
requirements for "reforming" methanol with steam); the onboard reformer is an added complication to the
system; and a methanol fuel cell vehicle would not qualify as a zoo-emission vehicle, because of the modest air
pollution emissions associated with the operation of the methanol reformer—although its emissions would be
much less than for an internal-combustion-engine vehicle. On balance, the methanol fuel cell vehicle may be
slightly less costly on a lifecycle cost basis than the hydrogen fuel cell vehicle [6,32],
The ability of the biomass fuels to compete arises from the fact that both hydrogen and methanol can be
used in technologically superior fuel cell vehicles, but gasoline and other hydrocarbon fuels cannot-at least for
first-generation fuel-cell vehicles.1
3 Over the oe*t couple of decades the most likely candidate fuel cell lor automotive applications it the 10-called
proton-exchange-membrane fuel cell, which opc riles at a modest temperature of 25 to 120 *C. Al nidi a low operating temperature it is
practical to reform methanol fuel, but the costs involved in reforming other fuels would be prohibitive. In the longer term, if
high-temperature fuel cells (e.g., the solid oxide fuel cell, which would operate at 1000 *C) prove to be practical for vehicular applications, it
may be feasible to reform a wide range of hydrocarbon or alcohol fuels under the bood of the car.
6-9
-------
The prospect thai biomass-derived methanol used in fuel cell vehicles will be able to compete with
gasoline used in internal-combustion-engine vehicles implies that liquid-fuel importing countries should be
indifferent to the choice between oil and biomass-derived methanol imports on narrow economic grounds. And
since methanol derived from biomass and coal feedstocks should also be roughly competitive on narrow
economic grounds while biomass would be favored on environmental grounds, biomass-derived methanol could
become a major energy carrier in international commerce in a world that is sensitive to environmental values.
This is the basis for the world trade pattern for liquid fuels shown in Figure 4 for a renewables-intensive global
energy scenario [5], with trade levels equal for oil and biomass-derived methanol in the period 2025-2050.
CREATING MAJOR ENERGY ROLES FOR BIOMASS WITH LIMITED LAND RESOURCES
Because the photosynthetic process is a relatively inefficient way of converting solar energy into
chemical fuel energy, large land areas are required if biomass is to make major contributions to energy supply.
For example, displacing fossil fuels in the US with the energy equivalent amount of biomass grown on
plantations at the average productivity of US forests (4 dry tonnes per hectare per year) would require a
plantation area of 1 billion hectares-approximately the total US land area. This "back-of-the-envelqpe"
calculation suggests that biomass can never become a significant energy source. While it is certainly true that
biomass resources are not large enough to enable biomass to provide all energy needs, the role of biomass
nevertheless can be substantial, if modem technologies are used for biomass production and conversion.
The land constraints on biomass production can be reduced in part by intensively managing the biomass
plantations. With modem production techniques, biomass productivities far in excess of natural forest yields can
be realized A reasonable goal for the average harvestable yield on large-scale plantations in the US is 15 dry
tonnes per hectare per year [3]~corresponding to a photosynthetic efficiency of about 0.5%. The amount of land
that might be committed to biomass energy plantations in the US could be over 50 million hectares--the amount
of excess cropland projected by the US Department of Agriculture to be be available by the year 2030 [14].
However, for the purposes of the present discussion, it is assumed that by that time the amount of land in the US
committed to biomass plantations is a more modest 30 million hectares-approximately the amount of excess
cropland in the US at present Potential biomass production on this much land at 15 tonnes per hectare per year
would amount to about 9 EJ per year.
The land constraints on biomass production can also be eased by exploiting for energy purposes biomass
residues (urban wastes and residues of the agricultural and forest-product industries) that can be recovered in
environmentally acceptable ways. It has been estimated that such residues in the US could amount to about 6 EJ
per year [28].
The biomass energy potentially available from these two sources, some 15 EI per year, could probably
be produced in the US in environmentally acceptable ways without running up against significant land-use
constraints. This is equivalent to 19% of total US primary energy use, exclusive of biomass, in 1987. It does
not follow, however, that these potential biomass supplies would displace 19% of conventional US energy. The
extent to which conventional energy would be displaced depends sensitively on the conversion technologies
deployed.
Consider, for example, the two energy activities often targeted for replacement by biomass energy-the
generation of electricity from coal and the running of light-duty vehicles (automobiles and light trucks) on
gasoline. In the US these activities in 1987 accounted for some 30 EJ of primary energy and about half of total
C02 emissions from fossil fuel burning. If these two activities (at 1987 activity levels) could be replaced by
biomass grown renewably, the result would be a 50% reduction in US COj emissions.
Suppose first that biomass were used with commercially available technologies: (i) replacing coal-based
steam-electric power plants with biomass-based steam^electric power plants having a 20% average efficiency and
(ii) replacing gasoline-fired internal-combustion-engine light-duty vehicles having 1987 average fuel economies,
with the intemal-combustion-engine vehicles operated on methanol derived from biomass (using commercially
available technology designed to make methanol from coal but modified to accommodate biomass), assuming no
4.
6-10
-------
improvement in the fuel economy of the vehicles other than what would be inherent in a shift from gasoline to
methanol (gasoline-equivalent fuel economies of 23 mpg for autos and 16 mpg for light trucks). The amount of
biomass needed for this conversion would be about 49 EJ per year (see Figure 5)--far more biomass than is
likely, to be available for energy purposes.
By the turn of the century, the first generation of biomass-integrated gasifier/gas turbine technology will
probably be commercially available, making it possible to roughly double the efficiency of biomass power
generation [33]. In ihis time frame more energy-efficient biomass-to-methanol conversion technologies may well
be available [32 ]. Moreover, it is feasible and cost-effective to introduce light-duty vehicles operated on
methanol having much higher fuel economies (gasoline equivalent fuel economies of 39 mpg for autos and 26
mpg for light trucks). Using these technologies the total biomass required to displace all coal power generation
and oil use by light-duty vehicles at 1987 activity levels would be reduced to 23.5 EJ per year (see Figure 5).
During the second decade of next century, even more energy-efficient fuel cell technologies are likely to
be available, both for power generation (57% efficient biomass-integrated gasifier/fuel cell systems employing
molten-carbonate or solid-oxide fuel cells) and for motor vehicle applications (proton-exchange-membrane fuel
cell vehicles that are 2.5 times as energy-efficient as comparable gasoline-fired intemal-combustion-engine
vehicles). Using these technologies the total biomass required to displace all coal power generation and oil use
by light-duty vehicles at 1987 activity levels would be reduced to about 14 EJ per year, which is comparable to
the above estimate of potential supplies from plantations and residues (see Figure 5).
Thus with advanced technologies biomass can play major roles in the energy economy, despite the low
efficiency of photosynthesis.
CONCLUSION
As a potential energy source for modem energy economies, biomass is unusual in that biomass
production and use would interact with a far wider range of human activities than any other energy source. In
part this is due to the fact that biomass energy is rooted in photosynthesis, without which human life would not
be possible. Yet agriculture and forestry are also based on photosynthesis, and biomass energy seems to have
the potential for even stronger interactions with human society than either of these other activities.
If biomass energy systems are poorly managed, they could provide some societal benefits, but the
benefits may not outweigh the social costs involved. But if biomass is well managed, it is very likely that all the
major concerns people have about biomass energy could be turned into a wide range of net societal benefits.
Successfully developed, biomass energy could provide:
o competitively priced modem energy carriers for a substantial fraction of human energy service
requirements, if advanced conversion and end-use technologies are used,
o the opportunity to reduce CO, emissions at zero incremental cost, through the displacement of fossil
6-11
-------
fuels by competitive biomass energy,4
o the opportunity to introduce biofuels that are compatible with zero-emission or near-zero-emission
fuel-cell vehicles, for combatting urban air pollution problems,
o the opportunity to bring competition, price stability, and energy security to the world fuels market
through the development of a large-scale world biofuels industry,
o a strong basis for rural development in developing countries,
o the opportunity to pay for the restoration of many tropical and subtropical degraded lands through their
conversion to biomass plantations for energy,
o the opportunity for sub-Saharan Africa and Latin America to become major world economic powers as
large-scale biofuels exporters,
o _a new livelihood for farmers in industrialized countries, and
o the opportunity to phase out agricultural subsidies in the industrialized countries, thereby strengthening
the economies of these countries while leveling the playing field in world trade for farmers in
developing countries.
But these benefits cannot be realized without a new public policy that promotes and coordinates
activities aimed at:
o eliminating the biases in current policies (subsidies, tax incentives, regulations) against biomass energy
systems,
o learning how best to produce biomass for energy sustainably under a wide range of conditions, while
preserving biological diversity and respecting a wide range of other environmental values,
o revitalizing and creating new economic opportunities for rural regions where biomass would be grown
for energy applications,
o establishing new biomass energy industries that can both produce and market modern biomass energy
carriers efficiently,
o carrying out the research, development, and demonstration needed to facilitate a continuing flow to the
market of innovative biomass energy conversion and end-use technologies,
o transferring resources needed to ensure access to advanced biomass energy technologies in developing
4 Id addition, tone cartoon would be sequestered is the iteady-state inventory of btomus piuumiicci Neglecting changes in toil
cartoon aisociaicd with establishing plantations tod considering only the avenge inventory associated with biomass that will be harvested, ihe
sequestering capacity of 400 million hectares of plantations would be about 9 billion tonnes of carbon (assuming an avenge rotation length
of 6 yean between cuts), corresponding to 3 yean of buildup of COj in the ionosphere from fossil fuel boning.
Changes in toil carbon should also be taken into account. Sane recent experimental evidence suggests that the sail carbon
content will increase at avenge rates in excess of one tonne of carbon per hectare per year during the fiirt IB to 20 yean of an unharvested
hybrid poplar stand established on land previously managed for row craps; this was detennined by comparisons of the carbon contents of the
soils in the hybrid poplar stand and the adjacent land in row crops. The experimental data are inadequate, however, to determine how long
this buildup might penist or the extent to which harvesting at intervals of about 6 yean would affect tbe nte of carbon turnover in the soil
[351.
6-12
-------
countries,
o establishing environmental guidelines that both give impetus to the development and commercialization
of clean biomass energy technologies and safeguard against environmentally undesirable approaches to
biomass energy development
The set of required new measures is broad, reflecting the broad range of benefits that could flow from successful
biomass energy development
Five years ago, the required policy change would have been deemed unrealitic, even unthinkable. But
today this is no longer the case. A global consensus is emerging that .the only acceptable development is
sustainable development This was the underlying theme of the United Nations Conference on Environment and
Development in Rio de Janeiro, in June 1992. It is now realized that the only way to achieve sustainable
development is to shift from one-dimensional policymaking to holistic approaches that deal with all direct and
indirect impacts of a given economic activity, making concerted efforts to avoid adverse impacts before they
occur.
Thus, though it is unfamiliar or at least not yet well understood by most people, biomass energy will get
focussed attention in the forthcoming sustainable development dcbalcs-both because of potentially disastrous
consequences of ill-planned biomass energy development efforts and because of the enormous overall benefits in
support of sustainable development goals that would arise from the proper development of biomass energy.
The timing of these debates could not be better for the biomass energy community. Because
modernized biomass energy plays a negligible role in the world energy economy at present, the future shape of
the biomass energy industry can be molded by the sustainable development debates, before the industry becomes
well established.
It is a rare event in modem history that the big concerns about potential adverse impacts of human
technological prowess are aired before people have had a chance to demonstrate this prowess by changing the
world. And rarer still is the possibility that these concerns can be reflected in timely changes of course that
avoid the problems of concern.
ACKNOWLEDGMENTS
This research was supported by the Office of Energy and Infrastructure of the US Agency for
International Development, the US Environmental Protection Agency's Air and Energy Engineering Research
Laboratory, the Geraldine R. Dodge Foundation, the Energy Foundation, the W. Alton Jones Foundation, the
Merck Fund, the New Land Foundation, and the Rockefeller Foundation.
6-13
-------
Table 1. Present4 and Potential6 Cropland for 91 Developing Countries (million hectares)
Present
Cropland
Potential Cropland
Region
Low
Rainfall
Uncer-
tain
Rainfall
Good
Rainfall
Natural
Flooded
Problem
Land
Desert
Total
C. America
37.6
2.2
13.3
18.5
5.7
31.4
3.5
74i6
S. America
141.6
26.0
37.5
150.3
105.7
492.7
2.8
815.0
Africa
178.8
73.4
96.8
149.3
71.3
358.1
3.8
752.7
Asia (excl.
China)
348.3
59.8
67.0
67.4
80.5
117.6
20.3
412.5
Total
706.3
161.4
214.7
385.5
263.1
999.7
30.4
2054.9
* Source: [13].
' As estimated by the Food and Agriculture Organization (FAO) in 1990 [19]. Potential cropland is defined
by the FAO as all land that is physically capable of economic crop production, within soil and water
constraints. It excludes land that is too steep or too dry or having unsuitable soils.
Table 2. Geographical Distribution of Tropical Degraded Lands and Potential Areas for Reforestation1
(million hectares)
Region
Logged Forests
Forest Fallow
Deforested
Watersheds
Desertified
Drylands
Total
Latin America
44.0
84.8
27 2
162.0
318.0
Africa
39.0
59.3
3.1
740.9
842.3
Asia
53.6
58.8
56.5
748.0
916.8
Total
136.5
202.8
86.9
1650.9
2077.1
Area Suitable
for
Reforestation
137
203
87
331
758
Notes to Tabic 2
¦ Source: [10].
6-14
-------
Table 3. Energy Balances for Biomass Production on Plantations*
Hybrid Poplar
Sorghum
Switchgrass
1990
2010
1990
2010
1990
2010
Energy
Inputs
(GJ/hectare)
Establishment
0.14
0.14
1.29
1.29
0.39
0.39
Fertilizers
3.33
3.33
8.87
12.69
5.26
7.38
Herbicides
0.41
0.41
1.82
1.82
-
-
Equipment
0.17
0.17
-
-
-
-
Harvesting
7.31
11.69
3.72
8.24
5.47
8.41
Hauling'
2.40
3.07
3.81
6.90
2.79
3.60
Total
13.8
18.8
19.5
30.9
13.9
19.8
Energy
Output
(GJ/hectare)
223.7
366.3
232.8
528.5
157.5
252.0
Energy Ratio
16.3
19.5
11.9
17.1
11.3
12.7
Notes to Table 3
• Source: [31).
b The energy required to transport the biomass 40 km to a biomass processing plant.
c Yields net of harvesting and storage losses for 1990 (2010) production technology are assumed to be
11.3 (18.5) tonnes per hectare per year for hybrid poplar (with a heating value of 19.8 GJ/tonne), 13.3 (30.2)
tonnes per hectare per year for sorghum (heating value of 17.5 GJAonnne), and 9.0 (14.4) tonnes per hectare
per year for switchgrass (heating value of 17.5 MJ/tonne).
4 The energy ratio = energy output/energy input.
-------
Nuclear 4.1
Biomass 14.7%
World
total * 373 exapules
Population = 4.87 bil'ion
Fneigy use per capita = 77 gigapules
Natural gas 22.TA
Stt. Nuclear 5.9S
Hydro S.7«,t
Biomass 2.8%
Industrialized countries
Total = 247 exajoules; 66 percent of world tota'
Population = t.22 billion; 25 percent of world total
Energy use per capita = 202 oigajoules
Oil 25.B*/i
Natural gas 7.1%
Nuclear 0.6 'A
Coal 23.4% Hydro S.1%
Biomass 38.1 S
Developing countries
"total =126 ensioules. 34 percent ol world total
Population = 3 65 billion; 75 percent of world total
Energy use oe' capita ¦ 35 gigajotries
Figure 1. World Primary Energy Consumption by Energy Source and by World Region
Primary energy consumption is shown for the world (top), industrialized countries (middle), and
developing countries (bottom) in 198S. Data for all energy sources except biomass are from [5).
Biomass energy data are estimates based on surveys, from [3].
The primary energy associated with electricity produced from nuclear and hydroelectric sources
is assumed to be the equivalent amount of fuel required to produce that electricity, assuming die average
heat rate (in MJ per kWh) for all fuel-fired power-generating units.
6-16
-------
Primary Energy Use for the Renevvables-intensive Global Energy Scenario
03
Q)
>-
Q>
Q.
ID
700
60D
500
100 -
300
| ) = CO? Fmissions (Reial've to 1983 = 1DD)
M00]
200
100
|C0 1]
S •'»'s S '/s'y ////¦
// , //, s, //.
AVlvV/XiwX'iv
(7V ,\r.W
AV/.KfcVViVA'/v*
V^v7X'>.VAVA'
~
19B5 2025 2050
Coal D Oil Q Natural Gas Nuclear
Hydro R Geothermal G3 Biotwss inleinrillent Renewable® H2 From Inteimitlenl Renewable*
Figure 2. Global Primary Energy Requirements far a Renewables-In tensive Global Energy Scenario
This figure shows global primary energy requirements for the renewables-intensive global energy
scenario developed in [5] in an exercise carried out to indicate the future prospects for renewable energy
for each of 1 i world regions. In developing this scenario, the high economic growth/high energy
efficiency demand projections for solid, liquid, and gaseous fuels and electricity developed by the
Response Strategies Working Group of the Intergovernmental Panel on Climate Change [36] were
adopted in [S] for each world region. For each region a mix of renewable and conventional energy
supplies was constructed in [5] to match these demand levels, taking into account relative prices, regional
endowments of conventional and renewable energy sources, and environmental constraints.
The primary energy associated with electricity produced from nuclear, hydroelectric, geothermal,
photovoltaic, wind, and solar thermal-electric sources is assumed to be the equivalent amount of fuel
required to produce that electricity, assuming the average heat rate (in MJ per kWh) for all fuel-fired
power-generating units in a given year. This global average heat rate is 8.05 MJ per kWh in 2025 and
6.65 MJ per kWh in 2050.
For biomass-derived liquid and gaseous fuels the primary energy is the energy content of the
biomass feedstocks delivered to the biomass energy conversion facilities.
Primary energy consumption in 1985 includes 50 EJ of non-commercial biomass energy [3J. It
is assumed that there is no next-commercial energy use in 2025 and 2050.
6-17
-------
Primary Energy Use for the Renewables-Intensive U.S. Energy Scenario
100
11 = C02 Emissions (Relative lo '985 = 100)
|39 4)
¦My//#//////''/,
,V % •,,.VvV> Vv*
^-z-;%vyM*yyys--i
|25 5|
-\w
19BS
2050
2025
| Coal B Oil ~ Natural Oas 0 Nuclear
| | Hydro 8 Geothermal [3 Blomasj ^3 Intermittent Bpnewab'es 0 From InlermiTtenl Renewable;
Figure 3. Primary Energy Requirements for the US in a Renewables-Intensive Global Energy Scenario
This figure shows primary energy requirements for the US in the rcnewables-intensive global
energy scenario developed in [5] in an exercise carried out to indicate the future prospects for renewable
energy for each of 11 world regions, one of which is die US. In developing this scenario, the high
economic growth/high energy efficiency demand projections for solid, liquid, and gaseous fuels and
electricity developed by the Response Strategies Working Group of the Intergovernmental Panel on
Climate Change [36] were adopted in [5]. For the. US and other industrialized countries, this demand
scenario involves a slow decline in primary energy demand as the economy expands, as a result of the
emphasis given to improved energy efficiency. The mix of renewable and conventional energy supplies
shown was constructed in [S] to match these demand levels, taking into account relative energy prices,
endowments of conventional and renewable energy sources, and environmental constraints.
The primary energy associated with electricity produced from nuclear, hydroelectric, geothermal,
photovoltaic, wind, and solar thermal-electric sources is assumed to be equivalent to the amount of fuel
required to produce that electricity, assuming the average heat rate (in MJ per kWh) for all fuel-fired
power-generating units in a given year. The US average heat rate is 8.07 MJ per kWh in 2025 and 6.42
MJ per kWh in 2050.
For biomass-derived liquid and gaseous fuels the primary energy is the energy content of the
biomass feedstocks delivered to the biomass energy conversion facilities.
6-18
-------
Interregional Fluid Fuel Exports in
Renewables-lntensive Global Energy Scenario
(millions of barrels of oil-equivalent per day)
Middle Easl
Other
Former CP Europe
/
Soulh and East Asia
Otter
North AlnCA
Middte East
Latin America Canada •
Alrica
Sub-Saharan Africa
Middle East
Other
Middle East
—Canada
Latin America
1985
(18.6 MMB/D)
[H3011
m Methanol from Biomass
SI Natural Gas
I I Solar Hydrogen
2050
(32.3 MMB/D)
Figure 4. Interregional Fuels Flows for a Renewables-lntensive Global Energy Scenario
The importance of world energy commerce for the renewables-intensive global energy scenario
developed in [5] and for which global primary energy consumption is shown in Figure 2 is illustrated
here. The figure shows that by the middle of the next century there would be comparable interregional
flows of oil, natural gas, and biomass-derived methanol, as well as small flows of hydrogen derived from
renewable sources. This diversified supply mix is in sharp contrast to the situation today, where oil
dominates international commerce in liquid and gaseous fuels.
Most methanol exports would originate in sub-Saharan Africa and in Latin America, where there
are vast degraded areas suitable for revegetation that will not be needed fa- cropland (see Tables 1 and
2), Growing biomass on such lands as feedstocks for producing methanol (or other biomass fuels) would
provide a powerful economic driver for restoring these lands.
6-19
-------
wgyg&rgtxj Biwness Residues
Biomass Plantations
Light Duty Vehtrles
Coal Based Power Generation
Actual Fossil Prsssnl Near-Tefro AtJvanced
Fuel Use Biomass Required to Replace Fossil Fuel
(1987) Using Alternative Technologies
Present: Sle»m-Eleclric Power. MeOH In ICE Vehicles (1987 Fuel Economy)
Neii-Term: BIC5'GT Power. MeOH In ICE Vehicles (Improved Fuel Economy)
Future. BIG/MoRen Carbonate Fuel Cell Power, MeOH In PEW Fuel Cell Vehicles
Potential
Biomass Supply
Figure S. Energy for Light-Duty Vehicles and Bower Generation in the US
Shown here are the biomass primary energy requirements for displacing all petroleum used by
light-duty vehicles (automobiles and light trucks) and all coal-fired power generation in the US, at the
1987 activity levels, with alternative biomass technologies, in relation to potential biomass supplies.
Hie bar on the left shows fuel actually consumed in 1987 by light-duty vehicles and by
coal-fired power plants. The second bar shows the biomass primary energy requirements if light-duty
vehicles and coal-fired power plants at 1987 activity levels were replaced by biomass energy systems that
are commercially available today. The third bar shows biomass requirments if technologies likely to be
available in the year 2000+ time frame were used to replace all oil used for light-duty vehicles and all
coal-based power generation, at 1987 activity levels, TTie fourth bar shows the biomass requirments if
technologies likely to be widely available in the 2020 lime frame were used to replace all oil used for
light-duty vehicles and all coal-based power generation, at 1987 activity levels. The bar on the right
shows potential biomass supplies from plantations on 30 million hectares of excess agricultural lands plus
residues (urban refuse plus agricultural and forest product industry residues) that are recoverable under
environmental constraints.
For details see Endnote for Figure 5.
6-20
-------
ENDNOTE FOR FIGURE 5
The bar on the left represents fuel consumed in 1987 by light-duty vehicles and by coal-fired power
plants. Automobiles and light trucks, with average fuel economies of 19.1 mpg and 12.9 mpg, respectively,
consumed 103 billion gallons of gasoline. In 1987 coal-fired power plants, operated with an average efficiency
of 32.9%, produced 1464 TWh of electricity.
The second bar shows the biomass primary energy requirements if light-duty vehicles and coal-fired
power plants at 1987 activity levels were replaced by biomass energy systems that are commercially available
today. With present biomass gasification technology (adapted directly from coal gasification) methanol can be
produced from wood at 50% efficiency (HHV basis). Operated on methanol, cars and light trucks would have
gasoline-equivalent fuel economies of 22.9 mpg and 15.5 mpg, respectively, some 20% higher than gasoline
vehicles, because of the higher thermal efficiency of internal combustion engines when operated on methanol
[29]. The net result is that the biomass feedstock requirements to support the 1987 level of light-duty vehicles
would be 1/(0.5*1.2) = 1.67 times the amount of gasoline used by light-duty vehicles in 1987. The present
average efficiency of biomass power plants operating in California is about 20%, so that the biomass plants
would require 32.9/20 = 1.65 times as much fuel to make electricity as the coal plants they would displace.
The third bar shows the biomass primary energy requirments if biomass technologies likely to be
available in the year 2000+ time frame were used to replace all oil used for light-duty vehicles and all coal-based
power generation, at 1987 activity levels. It is cost-effective to increase the average (on-the-road) fuel economy
of new cars and light trucks to about 33.6 and 22.7 mpg (76% higher than in 1987), respectively, if operated on
gasoline, and to 40.3 and 27.2 mpg of gasoline-equivalent energy (20% higher than on gasoline), respectively, if
operated on methanol. Such a shift could be achieved over the next couple of decades. During this period it
would be feasible to introduce methanol production technology involving indirect biomass gasification, for which
the overall biomass-to-methanol conversion efficiency is 63% [32]. With these technologies biomass fuel input
requirements would be (l/1.76)/(0.64* 1.2) = 0.74 times as large as the petroleum required in 1987. If electricity
were produced from biomass using biomass integrated gasifier/gas turbine technology that could be introduced
commercially in this time frame, the efficiency of power generation would be 39%, nearly double that of existing
biomass power plants, so that the biomass plants would require 32.9/39 - 0.84 times as much fuel to make
electricity as the coal plants they would displace.
The fourth bar shows the biomass primary energy requirments if biomass technologies likely to be
available in the 2020 time frame were used to replace all oil used for light-duty vehicles and all coal-based
power generation, at 1987 activity levels. By the end of the second decade of the 21st century, biomass-derived
methanol could be routinely used in proton-exchange-membrane fuel-cell vehicles at gasoline-equivalent fuel
economies that are 2.5-times the fuel economies of gasoline-powered internal-combustion-engine vehicles of
comparable performance [5, 6], With these technologies biomass fuel input requirements would be
(l/1.76)/(0.64*2.5) = 0355 times as large as the petroleum required in 1987. Also, by that time, biomass
integrated gasifier/fuel cell systems (using molten carbonate or solid oxide fuel cells) for stationary power
generation could well be available with biomass-to-electricity conversion efficiencies of perhaps 57% [33], for
which fuel requirements would be 32.9/57 = 0.577 times coal requirements for power generation in 1987.
The fifth bar shows potential US biomass supplies, consisting of (i) 9 EI per year of plantation biomass
grown (Hi 30 million hectares of excess agricultural lands at an average productivity of 15 dry tonnes per hectare
per year and a heating value of 20 GJ per dry tonne of biomass, plus (ii) 6 EJ per year of those residues (urban
refuse, agricultural residues, forest product industry residues) that are estimated to be recoverable under
environmental constraints [28].
6-21
-------
APPENDIX A: PHASING OUT AGRICULTURAL SUBSIDIES IN INDUSTRIALIZED COUNTRIES
BY CONVERTING EXCESS AGRICULTURAL LANDS TO BIOMASS PRODUCTION FOR ENERGY
Public policies provide subsidies to agricultural producers both via monetary transfers from taxpayers
through government budgets and from consumers through higher prices for agricultural commodities. These
subsidies are large-amounting in 1987 to S295 billion for all countries of the Organization for Economic
Cooperation and Development (OECD) and to $81 billion (some $330 per capita) for the US alone
[Ill-equivalent to about one-fifth of total retail expenditures on energy in the US [37]—see Table Al.
These subsidies were a focus of the May 1987 ministerial level meeting of the Council of the OECD.
The communique issued at this meeting declared [11]:
"Boosted by policies which have prevented an adequate transmission of market signals to farmers,
[agricultural] supply substantially exceeds effective demand. The cost of agricultural policies is
considerable, for government budgets, for consumers, and for the economy as a whole. Moreover,
excessive support policies entail an increasing distortion of competition on world markets: ran counter to
the principle of comparative advantage which is the root of international trade and severely damage the
situation of of many developing countries. This steady deterioration...cTeates serious difficulties in
international trade, which risk going beyond the bounds of agricultural trade alone...This deterioration
must be halted and reversed..."
The communique included as recommended agricultural policy reform principles:
"a) The long-term objective is to allow market signals to influence by way of a progressive and concerted
reduction of agricultural support, as well as by all other appropriate means, the orientation of
agricultural production: this will bring about a better allocation of resources which will benefit
consumers and the economy in general."
"0 The adjustment of the agricultural sector will be facilitated if it is supported by comprehensive policies
for the development of various activities in rural areas. Farmers and their families will thus be helped
to find supplemental or alternative income."
Despite the commitment to reform made in 1987, as of 1991 there was no fundamental reform of agricultural
policy for the OECD as a whole, as shown by time series data (see Table Al).
This stalemate could be ended and both of the above reform principles of the OECD ministerial
communique could be satisfied if policies were adopted to convert excess agricultural lands to biomass
production for energy purposes.
Principle T could be satisfied by eliminating the bias in current policies against energy crops. (In
many countries public policy specifies what crops the farmer can produce to qualify for a subsidy; and energy
crops are not allowed.) As the energy crop production industry becomes well established it would be feasible to
begin phasing out many of the subsidies, thus satisfying principle "a."
6-22
-------
Table Al. Transfers to Agricultural Producer! Associated with Agricultural Policies of OECD Countries*
1987
1988
1989
1990
1991
Ak
B*
Ak
B"
A"
Bc
Ak
r
A"
B'
Australia
0.6
38
0.6
35
0.7
39
1.1
67
1.2
70
Austria
3 8
501
3J
465
2.9
379
4.1
532
4.1
534
Canada
8.6
335
8.7
334
8.3
315
9.8
369
9,5
353
EC-12
119.4
369
120.8
372
101.2
311
138.3
401
141.8
409
Finland
4.4
894
4.9
997
4.9
994
6.0
1208
5.9
1173
Japan
65.5
537
70.7
577
65.8
534
60.4
489
63.2
510
N. Zealand
0.1
35
0.2
57
0.1
29
0.1
25
0.1
19
Norway
3.3
793
3J
825
3.3
790
4.2
993
42
987
Sweden
3.2
377
3.0
357
3.1
362
3 S
404
3.6
. 416
Switzerland
5.3
803
5.6
842
5.0
750
62
916
6.4
925
US
from
taxpayer*
51.6
44.2
47.8
44.2
51.8
from
consumer*
30.8
26.0
23.7
29.7
29.9
Tariff
revenues
- 1.4
- 1.1
-0.7
-0.9
- 1.0
Subtotal
80.9"
332
693d
281
70.8*
285
73.0
290
80.8
318
OECD
from
taxpayers
119.6
121.8
119 J
129.4
1429
from
consumer!
190.6
187.6
161.0
190.4
195.4
Tariff
revenues
• 15.2
- 18.6
- 14.4
- 13.1
- 17.6
Total
295.2
361
290.7
353
266.1
321
306.8
367
320.7
380
• Source: fill.
k A = toul transfers. in billion US dollar* per year.
' 8 = transfer! per capita, in US doUan per capita per year.
4 For companion, retail expenditure! on energy in the US ir»al]f4 S394 billion in 1987. $408 billion in 1988, and $437 billioa in 1989
137],
6-23
-------
CITATIONS
1. Michael Williams, Americans and Their Forests: a Historical Geography, Cambridge University Press,
Cambridge, England, 1989.
2. Ged R. Davis, "Energy for Plana Earth," Scientific American, vol. 263, no. 3, pp. 54-63, 1990.
3. David O. Hall, Frank Rosillo-Calle, Robert H. Williams, and Jeremy Woods, "Biomass for Energy: Supply
Prospects," pp. 593-652, in Renewable Energy: Sources for Fuel and Electricity, T. B. Johansson, H. Kelly, A.
K. N. Reddy, and R. H. Williams, eds., Island Press, Washington, DC. 1992.
4. Matthew L. Wald, "California's Pied Piper of Clean Air," Business Section, The New York Times, 13
September 1992.
5. Thomas B. Johansson, Henry Kelly, Amulya K. N. Reddy, and Robert H. Williams, "Renewable Fuels and
Electricity for a Growing World Economy: Defining and Achieving the Potential," pp. 1-72, in Renewable
Energy: Sources for Fuel and Electricity, T. B. Johansson, H. Kelly, A. K. N. Reddy, and R. H. Williams, eds.,
Island Press, Washington, DC, 1992.
6. Robert H. Williams, "Fuel Cells, Their Fuels, and the US Automobile," Proceedings of The First Annual
World Car Conference, 20-24 June 1993, University of California at Riverside (in press).
7. C. D. Masters, D. H. Root, and E. D. Attanasi, "World Oil and Gas Resources-Future Production Realities,"
Annual Review of Energy, vol. 15, pp. 23-31, 1990.
8. Intergovernmental Panel on Climate Change, The IPCC Scientific Assessment, J. T Houghton, G. J. Jenkins,
and J. J. Ephraums, eds., Cambridge University Press, Cambridge, England, 1990.
9. Jose Goldemberg, Thomas B. Johansson, Amulya K. N. Reddy, and Robert H. Williams, Energy for a
Sustainable World, 517 pp., Wiley-Eastern, New Delhi, 1988; and J. Goldemberg, T. B. Johansson, A. K. N.
Reddy, and R. H. Williams, Energy for a Sustainable World, 119 pp.. World Resources Institute, Washington,
DC, 1987.
10. A. Grainger, "Estimating Areas of Degraded Tropica] Lands Requiring Replenishment of Forest Cover,"
International Treecrops Journal, vol. 5, pp. 1-2, 1988.
11. Organization for Economic Co-Operation and Development, Agricultural Policies, Markets, and Trade:
Monitoring and Outlook 1992, Paris, 1992.
12. Kirk R. Smith and Susan A. Thorneloe, "Household Fuels in Developing Countries: Global Warming,
Health, and Energy Implications," in Proceedings from the 1992 Greenhouse Gas Emissions and Mitigation
Research Symposium, sponsored by the US Environmental Protection Agency, Air and Energy Engineering
Laboratory, Washington DC, 18-20 August 1992.
13. World Resources Institute, World Resources 1992-93, A Guide to the Global Environment, Oxford
University Press, 1992.
14. USDA-Soil Conservation Service, The Second RCA Appraisal: Soil, Water, and Related Resources on
Nonfederal Land in the United States-Analysis of Condition and Trends, June 1989.
15. New Farm and Forest Products Task Force, New Farm and Forest Products: Responses to the Challenges
and Opportunities Facing American Agriculture, a report to the Secretary of Agriculture, 25 June 1987.
6-24
-------
16. F. C. Hummel, "Biomass Forestry: Implications for Land-Use policy in Europe," Land Use Policy, pp.
375 384. October 1988.
17. Netherlands Scientific Council for Government Policy. Grounds for Choice: Four Perspectives for the
Rural Areas In the European Community, The Hague, The Netherlands, 1992
18. Intergovernmental Panel on Climate Change (World Meteorological Organization/United Nations
Environment Program), Climate Change: The IPCC Response Strategies, Island Press, Washington, DC, 1991.
19. Food and Agriculture Organization of the United Nations, Agricultural Land Use: Inventory of
Agroecological Zones Studies, FAO, Rome, 1991
20 Food and Agriculture Organization of the United Nations. Forestry In China, FAO Forestry Paper 35,
Rome, 1982.
21. Office of Technology Assessment of the U.S. Congress. Technologies to Sustain Tropical Forest Resources
and Biological Diversity, OTA-F-515, Washington, DC. U.S. Government Printing Office, May 1992.
22 Walter E. Parham, Patricia J. Durana, and Alison L. Hess, eds., Improving Degraded Lands: Promising
Ezerlences from South China, Bishop Museum Bulletin In Botany 3, The Bishop Museum, Honolulu. Hawaii.
1993.
23 "The Noordwijk Declaration on Atmospheric Pollution and Climatic Change," Ministerial Conference on
Atmospheric and Climauc Change, Noordwijk. The Netherlands, November 1989.
24. D. S. DcBell. C. D. Whitesell. and T H. Schubert, "Using N,-Fi*ing AIbain to Increase Growth of
Eucalyptus Plantations in Hawaii," Forest Science, vol. 35. no. I. pp. 64-75. I9B9.
25. Jcrald L Schoor, "Soil as a Vulnerable Environmental System," paper presented at the 1992 Global Change
Institute: Industrial Ecology and Global Change, Snowmass, CO, 20-31 July 1992.
26. S. Under, "Nutrient Control of Forest Yield," in Nutrition of Trees, Marcus Wallenberg Foundation
Symposia, vol. 6, pp. 62-87. Falun. Sweden, 1989.
27. J. P. Kimmins. "Modelling the Sustainability of Forest Production and Yield for a Changing and Uncertain
Future," The Forest Chronicle,vol. 6, pp. 271-280. June 1990.
28. J. Beyea. J. Cook, D. O Hall, R. H. Socolow. and R. H. Williams. Toward Ecological Guidelines for
Ijirge-Scaie Biomass Energy Development, Report of a Workshop for Engineers, Ecologists, and Policymakers
Convened by the National Audubon Society and Princeton University, 6 May 1991, National Audubon Society,
New York. 1992.
29. Charles E. Wyman, Richard L. Bain, Norman D. Hinman, and Don J. Stevens, "Ethanol and Methanol from
Cellulosic Biomass." pp. 865-924, in Renewable Energy: Sources for Fuel and Electricity, T. B. Johansson, H.
Kelly, A. K. N. Reddy, and R. H. Williams, eds., Island Press, Washington, DC, 1992.
30. Jose Gotdcmberg, Lourival C. Monaco, and Isiaias Macedo, "The Brazilian Fuel-Alcohol Program," pp.
841-864, in Renewable Energy: Sources for Fuel and Electricity, T. B. Johansson. H. Kelly, A. K. N. Reddy.
and R.H. Williams, eds.. Island Press, Washington, DC, 1992.
31. A. H. Turhollow and R. D. Per lack, "Emissions of CO, from Energy Crop Production," Biomass and
Bioenergy, vol. 1, pp. 129-135, 1991.
32. Ryan E. Katofsky. Production of Fluid Fuels from Bhmass, PU/CEES Report No. 279, Center for Energy
and Environmental Studies, Princeton University, Princeton, NJ, June 1993.
33. Robert H Williams and Eric D. Larson, "Advanced Gasification-Based Power Generation," pp. 729-786, in
Renewable Energy: Sources for Fuel and Electricity, T. B. Johansson, H. Kelly, A. K. N. Reddy, and R. H.
Williams, eds.. Island Press, Washington. DC. 1992.
34. Jane Tumbull, "Strategies for Achieving a Sustainable, Clean, and Cost-Effective Biomass Resource,"
Electric Power Research Institute, Palo Alto. CA. January 1993
35. L. L. Wright and E. E. Hughes, "U. S. Carbon Offset Potential Using Biomass Energy Systems," Journal of
Water, Air, and Soil Pollution fin press).
36 Response Strategies Working Group of the Intergovernmental Panel on Climate Change, "Emissions
Scenarios: Appendix of the Expert Group on Emissions Scenarios." US Environmental Protection Agency.
Washington, DC, April 1990.
37. Energy Information Administration, State Energy Price and Expenditure Report I9S9, DOE/EIA-0376<89),
September 1991.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's peer
administrative review policies and approved for presentation and publication.
6-25
-------
AN ANALYSIS OF THE HYDROCARD PROCESS
FOR METHANOL PRODUCTION FROM BIOMASS
by: Yuanji Dong and Meyer Steinberg
Hvdrocarb Corporation
232 West 40th Street
New York, NY 10018
and Robert H. Borgwardt
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
ABSTRACT
A pilot plant is being designed to evaluate the technical feasibility of producing
transportation fuel from biomass by the Hydrocarb process. As a basis for
that design, computer simulations and experimental studies have been carried out to estEblish
optimum process conditions for a range of feedstocks that are anticipated for pilot plant tests.
This paper discusses the results of simulations to determine the operating parameters and
performance when using urban wastes such as greenwaste and sewage sludge as feedstocks. The
simulations were used to configure the process steps for maximum fuel (methanol) production,
to determine feed rates, and to estimate thermal efficiency. The results indicate that about 77
kg of methanol can be produced from 79 kg of dry greenwaste when sludge and digester gas are
fed as co-feedstocks in a ratio of 0.2. The optimum system pressure is found to be 50 atm (5
MPa). Temperatures of 900"C for gasification and 1-000®C for methane pyrolysis are
recommended on the bases of thermodynamics, kinetics, and the limitations of materials of
construction. Thermal efficiency at these conditions is estimated to be 74 percent.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication.
6-26
-------
INTRODUCTION
Thermodynamic calculations suggest that the Hydrocarb process0-2' might produce carbon
black, hydrogen, and/or methanol from fossil fuels or virtually any other carbonaceous materials
and might do so with high thermal efficiency. Potential feedstocks include coals of all ranks,
residual oil, oil shale, woody biomass, sewage sludge, and municipal wastes. This innovative
process consists of three essential reactions: (1) hydrogasification of the carbonaceous feedstocks
to produce a methane-rich gas, (2) thermal decomposition of methane to carbon and a hydfogen-
rich gas which is recycled, and (3) catalytic conversion of the carbon monoxide (CO) and
hydrogen contained in the recycled gas to produce methanol. The mix of process products is
optional: a clean solid fuel (carbon black), a liquid fuel (methanol), slurry fuel mixtures of
carbon and methanol, or gaseous fuels (hydrogen and methane) can be produced by changing the
order of the reaction steps.
Since the process operates without additional steam and oxygen--and because the oxygen
in the feedstocks is removed mainly in the form of methanol and water-the carbon dioxide (CO^
emission is significantly reduced in comparison to traditional fossil fuel conversion processes
involving steam/oxygen gasification"'. When biomass is used as a co-feedstock with fossil fuel
to produce methanol and the carbon is sequestered, the C02 greenhouse gas emission can be
significantly reduced and even eliminated(0).
PRELIMINARY ANALYSES
In order to attain a better understanding of the potential of the Hydrocarb process,
particularly as a means of producing a clean transportation fuel from biomass, the U.S.
Environmental Protection Agency (EPA) developed an interagency agreement with the
Brookhaven National Laboratory to perform detailed process analyses and related experimental
studies of biomass hydrogasification and methane pyrolysis. The process was analysed using a
computer simulation model developed by Hydrocarb Corporation that performs complete mass
and energy balances for various process configurations, feedstock options, reactor types, and
operating conditions. Initial results of the simulations were published by the EPA in 1991(3>; a
follow-up report is in preparation.
EPA's independent assessment of the biomass/natural-gas option of the Hydrocarb process
as a technology for production of alternative transportation fuels'61 confirmed that it can,
theoretically, produce methanol at a cost that is competitive with petroleum fuels. Most
importantly, it concluded that methanol may be produced and utilized in the transportation sector
with a 70 percent reduction in CO, emission relative to gasoline at no incremental cost and may
achieve 100 percent C02 reduction at marginal incremental cost. Other potential advantages
derive from its higher yield of fuel energy compared to other processes for producing alternative
fuels from biomass or from natural gas by the conventional steam reforming process. The
Hydrocarb approach therefore has potential for mitigating C02 emissions from mobile sources
in a more effective manner than current alternatives. Since the transportation sector accounts for
at least 24 percent of total C02 emissions, any means of reducing that source will have to be
considered when assessing options for dealing with the global warming problem.
6-27
-------
PILOT PLANT EVALUATION
Because of methanol's potential role in reducing urban air pollution, the South Coast Air
Quality Management District (SCAQMD) of California is seeking new technologies for domestic
methanol production, especially from urban wastes, and has agreed to co-sponsor with the EPA
a pilot plant evaluation of the Hydnocarb process. Preliminary design of that pilot plant is now
in progress. The objectives are: (1) to demonstrate the technical feasibility of producing methanol
from woody biomass and natural gas by the Hydrocarb process, and (2) to evaluate its
performance using sewage sludge, digester gas, and greenwaste as alternative feedstocks. If
successful, the results will be used to establish more comprehensive cost estimates and provide
a basis for decisions regarding further development. Prior analyses (XS,6> have considered the use
of woody biomass and natural gas as feedstocks. This paper focuses on the design of a pilot
plant that will utilize 50 Ib/hr (23 kg/hr) of urban wastes as feedstock.
LABORATORY STUDIES
KINETICS OF BIOMASS HYDROGASIFICATION
The hydrogasification of biomass in the form of poplar wood having panicle size less than
150 pm in diameter was investigated in a 25 mm ID and 2.5 m long tubular reactor facility
described in detail elsewherea,). The tests were conducted at temperatures up to 800°C and
pressures between 30 and 50 atm. The experiments were performed in two modes depending
upon the heatup rate. In the low heatup mode, the biomass was first loaded in the reactor at
room temperature. Hydrogen was then introduced into the system until the desired initial
pressure was established. The reactor was slowly heated at a rate less than lCC/min while the
change of pressure in the reactor and the composition of the effluent gas were monitored with
time. In the higher heatup mode, the reactor was first heated to the desired temperature and
pressurized with hydrogen before introducing the biomass. From the variation of pressure and
gas composition versus time, the rates of reaction and degree of conversion were determined.
The high-heatup-rate experiments showed that 88 to 90 wt percent of biomass could be gasified
at residence times in the order of 15 minutes^. The number of moles of gas formed, as
calculated from the pressure change, varies with time as shown in Figure 1.
METHANE PYROLYSIS
The rate of thermal decomposition of methane was investigated using the same reactor
facility at temperatures ranging from 700 lo 900°C and pressures of 28 to 56 atm**'. In these
experiments, methane was continuously fed into the reactor. On-line gas analyses were taken
upstream and downstream of the reactor to determine the reaction rate. The variation of methane
concentration with residence time at different operating conditions, shown in Figure 2, indicates
that a gas residence time of about 2 min is required for the reaction to reach equilibrium
composition at 50 atm and 900°C. The activation energy is found to be 31.3 kcal/mol. By
extrapolating it to higher temperatures, the residence time would be 41 sec at 1000°C and 12 sec
at 1140°C.
6-28
-------
0 10 SO 30 40 60 BO
TIME (min.i
Figure 1. The change in number of moles in the reactor with lime at 800"C and 5.24 aim
of initial hydrogen pressure. Run No. 1152
o
o
o
¦2.
90
Z
Q
h-
80
<
oc
H
70
Z
UJ
o
60
z
o
o
UJ
50
z
<
E
40
UJ
2
JO
n r
Equilibrium
"concentration
—' ' '
0 20 40 60 80 100 120
RESIDENCE TIME (sec)
Figure 2. Methane concentration vs. gas residence time. (56.1 atm and 9CKTC)
6-29
-------
PROCESS SIMULATION STUDIES
REACTION SEQUENCE
There are two possible configurations for the Hydrocarb process, according to the
sequence of the three reactor steps involved. One of these configurations, Cycle 1, is designed
so that the process gas flows successively from the hydrogasification reactor (HPR) through the
methane pyrolysis reactor (MPR), the methanol synthesis reactor (MSR), and the methanol
condenser (COND). The gas leaving the condenser is recycled back to the HPR after
withdrawing a small purge gas stream. The other configuration, Cycle 2, directs the process gas
formed in the HPR to the MSR and condenser, then to the MPR, followed by recycle to the HPR.
Figures 3 and 4 illustrate these configurations.
Purga Gas MflOH
COND
Carbon
MPR
HPR
MSR
HE
Figure 3. Cycle 1 of the Hydrocarb process.
In Cycle 1, the process gas from the MPR is cooled from the temperature of that reactor
down to the MSR reaction temperature, 260°C, and this recovered energy is used to heat the
process gas from the condenser, operating at 50°C, up to about the temperature of the HPR.
From an energy balance on the HPR (assumed to be a fluidized bed), the temperature of the
recycled process gas fed into it is adjusted until the reaction heat generated in the HPR is
balanced by the enthalpy difference of inlet and outlet gas streams. In Cycle 2, two gas/gas heat
exchangers are used: one cools the gas stream from the temperature of the HPR to the
temperature of the MSR by heating the gas stream from the condenser; and before entering the
MPR, the process gas is further heated in the second heat.exchanger by the hot gas stream from
the MPR.
6-30
-------
Purge Ga# MeOH
Carbon
MSR
MPR
HE
HP R
HE
COND
Figure 4. Cycle 2 of the Hydrocarb process.
The primary interest of SCAQMD and EPA is to produce a maximum amount of
methanoJ from biomass. With this in mind, computer simulations were made with the process
model to compare the above cycles and determine which is better suited to that objective. The
feedstocks assumed for this comparison are representative of urban wastes obtained from a
California wastewater treatment plant (sewage sludge and digester gas) and other sources
(greenwaste). The compositions of these materials are listed in Table 1. Included in Table 1
are the higher heating values (HHVs) of each fuel, given on a moisture-free basis (MF), and the
heat of formation, which is expressed on a moisture-and-ash-free basis (MAF).
Performance Criteria
Several important performance criteria are used to compare the two cycles and also to
assess the performance of Hydrocarb with alternative feedstocks. They include: minimum
methane feed per unit of biomass feed, production ratio of methanol to carbon black, carbon
efficiency, thermal efficiency, and gas circulation rate. Thermal efficiency is defined here as the
total heating value of methanol, carbon black, and net purge-gas production (i.e., that which is
not used to heat the MPR) divided by the total heating value of biomass, natural gas, and any
external fuel source that is used to heat the MPR or close the energy balance. Similarly, carbon
efficiency is the ratio of total carbon in all of the products to the carbon in all of the feedstock
materials including the fuel used for MPR heating.
6-31
-------
TABLE 1. PROPERTIES OF THE FEEDSTOCKS USED
IN THE STUDY
Feedstock
Green-
waste
Methane
CH,
Sewage
Sludge
Digester
Gas
Molecular Formula
Composition (wt%)
C
H
O
H,0
Ash
S
N
CH,
CO,
High Heating Value
(Btu/lb-MF)*
(kcal/kg-MF)"
Heat of Formation
(kcal/kg-MAF)
CH1.4O0. si
49.11
5.85
35.41
5.00
3.34
0.16
1.13
-8670
-4817
-1386
CH,
100
-23881
-13267
cHj.7O0.4j ch}3o07I
28.55
4.09
16.03
9.82 -
36.53
1.36
3.62
36.81
61.82
-5510
-3061
-1770
-8792
-4884
' 1 Btu/lb = 2 kJ/kg
b 1 kcal/kg = 4 kJ/kg
Using ihe above performance criteria. Table 2 compares Cycles 1 and 2 under the same
operating conditions when the feedstocks are greenwaste and natural gas. The comparison is
based on a capacity of 100 kg/hr of greenwaste containing 5 wt percent moisture. The HPR Is
assumed to be a fluidized bed in which 90 percent of the carbon content of the biomass is
gasified (an assumption justified by the experimental work discussed above). The material
balances calculated for the two cycles show that a minimum natural gas feed equivalent to 10
wt percent of the biomass feed is required for Cycle 1 and 6 percent for Cycle 2.
TABLE 2. COMPARISON BETWEEN CYCLE 1 AND CYCLE 2
WITH BIOMASS AND NATURAL GAS AS FEEDSTOCKS
(50 atm, HPR ¦ 900"C, AND MPR • 1000°C)
Cycle 1 2
Greenwaste (5% H,0) (kg/h) 100 100
CH( Rate (kg/h) 10 10
Burning CH4 for MPR (kg/h) 12.8 11.6
Methanol (kg/h) 58.0 31.6
Carbon Black (kg/h) 29.7 35.6
MeOH/C (kg/kg) 1.95 0.89
Carbon Efficiency (%) 77.8 75.1
Thermal Efficiency (%) 72.0 64.4
Gas Recycle (kgmol/h) 46.4 43.6
6-32
1
-------
Table 2 compares the results of process simulations of the two cycles under ihe same
operating conditions using greenwaste and natural gas as co-feedstocks. It can be concluded that
Cycle 1 provides the higher methanol production and greater thermal efficiency; moreover, it
requires only one heat exchanger. Thus, Cycle 1 is the configuration that will be used for pilot
plant evaluation.
FEEDSTOCK SIMULATIONS
Greenwaste Moisture
As shown in Table 2, 100 kg of greenwaste containing 5 percent moisture should produce
58 kg of methanol when Hydrocarb is operated in the Cycle 1 configuration. The moisture
content of the feedstock was found to have an influence on process performance as illustrated
for greenwaste in Table 3. The basis for these calculations was 95 kg/h of moisture-finee
greenwaste to which varying amounts of water are added. The reaction temperature was 900°C
for the HPR and 1000°C for the MPR. The table shows that the minimum methane feed ratio,
which is the weight ratio of methane to greenwaste fed into the HPR for hydrogasification,
increases with moisture content of the greenwaste in order to maintain material balance in the
system. Table 3 also indicates that the methanol yield and the thermal efficiency improve with
increasing moisture content of greenwaste. However, as the moisture in biomass fed into the
HPR increases, the exothermic heat generated in the reactor decreases, resulting in the
requirement of a higher temperature of the process gas recycled into the HPR in order to keep
its reaction temperature constant.
TABLE 3. EFFECT OF MOISTURE IN GREENWASTE
ON THE PERFORMANCE OF THE PROCESS
(based on 95 kg/h MF greenwaste)
HjO/MF-GW (wt.%) 5.26 10.5 15.8 21.1
Min. CH,/MF-GW (kg/kg)
0.1
0.12
0.14
0.16
T of PG into HPR (eC)
892
910
927
944
Methanol (kg/h)
58.0
65.5
73.1
80.8
Carbon Black (kg/h)
29.7
28.4
27.0
25.5
MeOH/C (kg/kg)
1.95
2.31
2.71
3.17
Carbon Efficiency (%)
77,8
77.8
77.7
77.7
Thermal Effiency (%)
72.0
72.8
73.4
74 .1
Gas Recycle (kgmol/h)
46.4
47.5
48.6
49.8
When the weight ratio of moisture to greenwaste reaches 21.1 percent, the temperature
of the process gas recycled to the HPR is required to be 944°C. If the moisture of greenwaste
is further increased, the overall reaction heat in the HPR will become endothermic. Therefore,
there is an optimum moisture content of the feedstock for maximum methanol yield. For
greenwaste, that optimum is 21 percent (Table 3), which yields about 81 kg of methanol for
every 95 kg of dry biomass. Biomass moisture contents can be adjusted to the optimum value
by water addition or by external drying using waste heat from the MPR heater or from the
methanol converter.
6-33
-------
Sewaee Sludge and Digester Gas
Additional simulations were made assuming the feedstocks to be sewage sludge and
digester gas, supplemented by greenwaste and natural gas. These comparisons are based on a
fixed capacity of 100 kg/h of total solid feedstocks including sludge and greenwaste. For these
calculations, both the greenwaste and sludge were assumed to be dried to 5 percent moisture
before entering the HPR. The process performance of Cycle 1 at 50 atm and 900°C for the HPR
and 1000°C for the MPR is summarized in Table 4 assuming various feed ratios of sludge to
greenwaste. The methane feed was 15 percent of the greenwaste rate for all cases, while the
ratio of digester gas to sludge was 0.5, based on the gas generating capacity of the sewage plant.
A maximum allowable sludge feed ratio of 0.4 (sludge/greenwaste) was found in the calculation;
if this ratio is further increased, the HPR becomes endothermic and an additional heat source
would be needed for the HPR.
TABLE 4. THE EFFECTS OF SLUDGE/GREENWASTE FEED
RATIO ON THE PROCESS PERFORMANCE
<50 atm, HPR s 900°C, and MPR = 1000°C)
Sludge/Greenwaste 0 0.1 0.2 0.3 0.4
Greenwaste (Jcg/h)*
Sludge (kg/h)a
CH4 (kg/h)
Digester Gas (kg/h)
Limestone (kg/h)b
Methanol (kg/h)
Carbon Black (kg/h)
MeOH/C (kg/kg)
Carbon Efficiency (%)
Thermal Efficiency (%)
100.0
90.90
83.33
76.92
71.43
0
9.10
16.67
23.08
28.57
15.00
13.61
12.50
11.53
10.69
0
4.54
8.33
11.53
14.26
1.43
2.47
3.32
4.07
4.66
56.44
57.73
58. 92
59. 91
60.62
31.99
31.32
30.83
30.38
29.94
1.76
1.84
1.91
1.97
2.02
79.65
79.62
79.59
79.57
79.55
72.30
72. 65
72.94
73.19
73.40
* 5% moisture content in greenwaste and sludge.
b Molar feed ratio of limestone to total sulfur in
feedstocks is assumed to be 2.0.
The data in Table 4 show that the thermal efficiency and the product ratio are only
slightly affected by the sludge/greenwaste ratio. This indicates that sludge can be processed with
greenwaste by the Hydrocarb process to produce considerable amounts of methanol and carbon
black. Calculations also found that there is an optimum moisture content for each
sludge/greenwaste feed ratio. As shown in Table 5, increasing the moisture content of
sludge/greenwaste feeds increases the product ratio of methanol to carbon black and the thermal
efficiency of the process. Since the maximum allowable ratio of sludge to greenwaste is 0.4, the
pilot plant design took a sludge/greenwaste ratio of 0.2 as a representative data base. The
optimum flow rate and composition of each stream for the feed ratio (sludge/greenwaste) of 0.2
is shown in Figure 5. When feeding digester gas, it is interesting to note that part of the COj
component of that gas reacts in the HPR with carbon in the greenwaste to form CO and methane
and the rest of that C02 is converted to CO in the MPR. This results in an increase in the CO
6-34
-------
mole fraction of the process gas from 2.24 to 5.79 percent in the HPR and from 5.79 to 7.82
percent in the MPR. The CO is then converted to methanol in the MSR. Some waste C02, as
well as waste biomass, is thus recycled to a clean fuel.
TABLE 5. OPTIMUM MOISTURE CONTENT FOR
SLUDGE/GREENWASTE FEEDSTOCKS
Sludge/Greenwaste
0.2
0.4
CH< (kg/h)
Digester Gas (kg/h)
Limestone (kg/h)
Methanol (kg/h)
Carbon Black (kg/h)
MeOH/C (kg/kg)
Carbon Efficiency (%)
Thermal Efficiency (%)
Dry greenwaste (kg/h)
moisture (kg/h)
Sludge (kg/h)
moisture (kg/h)
79.16
14.08
15.04
1.72
12.50
8.33
3.26
76.86
25.85
2.97
77.88
74.17
67. 86
9.96
27.14
2.96
10.69
14 .26
4 . 66
73.90
26.35
2.80
78.28
74 .30
OPERATING CONDITIONS
The pressure and temperature at which the HPR and MPR are operated affect the
performance of the Hydrocarb process. With regard to the HPR, the temperature is established
by the requirement for desulfurization, especially in the case of sewage sludge, which is to be
partially accomplished by limestone addition. Given the C02 partial pressures in the HPR, the
active hydrogen sulfide sorbent, calcium oxide, can exist only at temperatures above 850°C. For
that reason, and the fact that gasification rate and carbon conversion increase with temperature,
an HPR temperature of 900°C is established.
Temperature
The effect of temperature in the MPR was investigated at 50 atm pressure when the HPR
temperature is fixed at 900°C. The results shown in Table 6 indicate that increasing the MPR
temperature will improve the methanol production and reduce the process gas circulation rate.
However, the temperature in the MPR is limited by the thermal and structural properties of the
materials of construction. When a MPR temperature of 950°C is assumed, the HPR is no longer
energy neutral: it becomes endothermic, For these reasons, 1000°C is the recommended
temperature for the MPR.
6-35
-------
Pressure
The influence of pressure is also presented in Table 6. The methanol production rate is
found lo increase with decreasing system pressure. However, as pressure decreases, the thermal
efficiency is reduced and the gas circulation rate (which affects capital cost of the plant) increases
significantly. Since catalytic methanol converters normally require at least 50 atm pressure, some
simplification is achieved if the other two process steps that produce the synthesis gas are
conducted at the same pressure.
From this study it is concluded that the most favorable operating conditions for co-
processing biomass and methane are 50 atm system pressure, and temperatures of 900°C in the
HPR and 1000°C in the MPR.
TABLE 6. EFFECTS OF PRESSURE AND TEMPERATURE ON
THE PROCESS PERFORMANCE (100
kg/h GW,
CH,/GW « 0.15,
HPR =
900°C)
p
T/MPR
MeOH/C
C Eff
Ther Eff
Recycle Gas
atm
°C
kg/kg
%
%
kgmol/h
50
1000
1.76
79.7
72.3
47.5
50
1050
2.31
78.1
72.3
34.4
50
1100
2 .65
77.3
72.3
27.4
50
1200
2.96
76.4
71.9
20.4
60
1000
1.55
80.9
72.7
44.7
50
1000
1.76
79.7
72.3
47.5
40
1000
2.06
77.9
71.2
53.7
30
1000
2.31
74.8
67.9
68.3
PILOT PLANT DESIGN
Based on the above process analysis and kinetics experiments, design specifications are
being developed for the HPR and MPR of a 50 Ib/h Hydrocarb pilot plant utilizing biomass
feedstocks. The energy balance simulations are based on fluidized bed reactors which will be
used for both the hydrogasification of biomass and the decomposition of methane. The reaction
temperature in the HPR is controlled by adjusting the temperature of the inlet process gas from
the gas heat exchanger which recovers the heat in the hot process gas from the MPR.
The heat required for the MPR is to be provided by circulating alumina particles between
the fiuidized bed MPR and a riser combustor. Other fuels such as methane or by-product carbon
are burned in the riser to heat the alumina particles. The hot particles then enter the MPR where
their sensible heat is transferred to the entering gas. The cooled panicles are then returned to
the riser combustor for reheat. Detailed design specifications for the HPR, MPR, reheater, and
other equipment for the pilot plant are being developed.
6-36
-------
Sewage
Sludge 17
Limestone 3,
CII4 12
DG
Greenwaste
kg/h
kg/h
kg/h
kg/h
Ash/Char
17 kg/h
DRYER
GW 93.2
(15.lt moist)
52.2 kinol/h
14.9 kg/h
7.2%
29.6
COHB
55.3
1400
1000
Methanol
76.9 kg/h
Water
14.2 kg/h
purge Gas
0.01 knol/h
Carbon Black
25.9 kg/h
57.7 kmol/h
CO 9.8%
CD2 0.4
CH4 20.1
H20 3.9
H2 65.9
48.9
kmol/h
CO 2.9%
C02 3.3
CH4 23.7
H20 0.1
H2 69<1
COND
50 C
52.1 kmol/h
2.7 %
3.1
CO
C02
CI!4 22.3
H20 1.6
H2 64.9
NeOH 5.4
260 C
Figure 5. Rates and compositions of various streams for the optimum conditions with
greenwaste (GW), sludge (SL), CH,. and digester gas (DG) as feedstocks
(50 atm, SL/GW = 0.2).
6-37
-------
SUMMARY AND CONCLUSIONS
The process design and analysis as well as experimental work on the kinetics of
hydrogasification of biomass and methane decomposition for a pilot plant using the Hydrocarb
process to produce methanol from biomass are reviewed. The hydrogasification kinetic
experiments showed an overall conversion of 88 to 90 wt percent of biomass in the HPR. The
gas residence time required for the reaction in the MPR to reach an equilibrium composition was
found to be 2 min at 50 atm and 900°C. Process simulations show that a process having the flow
sequence from the HPR to the MPR to the MSR to the condenser, produces more methanol and
provides a higher thermal efficiency than the alternative cycle. A minimum methane co-feed is
required for processing greenwaste, and depends on the moisture content of the greenwaste.
Sewage sludge and digester gas from a municipal sewage treatment plant can be co-processed
with greenwaste by the Hydrocarb process. The C02 in digester gas is convened to CO in the
HPR and MPR so that more methanol can be produced in the methanol synthesis reactor. The
optimum process conditions were determined to be a pressure of 50 atm in the total system and
temperatures of 900 and 1000"C in the HPR and MPR, respectively.
REFERENCES
(1) Steinberg, M. The Direct Use of Natural Gas (Methane) for Conversion of Carbonaceous
Raw Materials to Fuels and Chemical Feedstocks. In: Hydrogen Systems, volume 2,
Proceedings of the International Symposium, Beijing, China. Pergamon Press, Oxford,
1986. pp. 217-228.
(2) Steinberg, M. and E.W. Grohse. HYDROCARB-M,m Process for Conversion of Coals to
a Carbon-Methano! Liquid Fuel. BNL-43569, Brookhaven National Laboratory, Upton,
NY, 1989.
(3) Steinberg, M., E.W. Grohse, and Y. Dong. A Feasibility Study for the Coprocessing of
Fossil Fuels with Biomass by the Hydrocarb Process. EPA-600/7-91-007 (NTIS DE91-
011971), U.S. Environmental Protection Agency, Research Triangle Park, NC, 1991.
(4) Steinberg, M. Biomass and Hydrocarb Technology for Removal of Atmospheric COr
BNL-44410, Brookhaven National Laboratory, Upton, NY, 1991.
(5) Borgwardt, R.H., M. Steinberg, E.W. Grohse, and Y. Dong. Biomass and Fossil Fuel to
Methanol and Carbon via the Hydrocarb Process. Energy Biomass Wastes, 15: in press,
1991.
(6) Borgwardt, R.H. A Technology for Reduction of C02 Emissions from the Transportation
Sector. Energy Conversion & Management, 33, No. 5-8, pp. 443-449, 1992.
(7) Kobayashi, A. and M. Steinberg. Hydropyrolysis of Biomass. BNL-47158, Brookhaven
National Laboratory, Upton, NY, 1992.
(8) Kobayashi, A. and M. Steinberg. The Thermal Decomposition of Methane in a Tubular
Reactor. BNL-47159, Brookhaven National Laboratory, Upton, NY, 1992.
6-38
-------
ALTERNATIVE FUELS FROM BIOMASS
Charles E. Wyman
Alternative Fuels Division
National Renewable Energy Laboratory
1617 Cole Boulevard
Golden, Colorado 80401
ABSTRACT
Substitution of biofuels derived from cellulosic biomass for conventional fuels would
reduce the accumulation of carbon dioxide in the atmosphere and the possibility of global
climate change, improve our energy security and trade deficit, revitalize rural and farm
economies, and address urban air pollution and waste disposal problems. The major
fractions of biomass, cellulose and hemicellulose, can be broken down into sugars that can
be fermented into ethanol. Through technology advances for producing ethanol, the
projected cost at the plant gate has been reduced from about $3.60/gallon ten years ago to
$1.27/gallon, and opportunities have been identified to further drop the price to
$0.67/gailon, a price competitive with gasoline from oil at $25/barrel, within ten years.
Through anaerobic digestion, a consortium of bacteria can break down cellulosic biomass
to generate a medium-Btu gas that can be cleaned up for pipeline-quality methane. The
cost of this methane has been reduced to about $4.50/MBtu (106 Btu) for municipal solid
waste (MSW) feedstocks, and technology advances could drop the price to about
$2.00/MBtu. Algae could consume carbon dioxide from power plants and produce lipid oil
that can be converted into a diesel fuel substitute. Projected costs for this biodiesel have
dropped from almost $18.00/gallon to about $3.50/gallon now, with a target of Sl.OO/gallon.
Biomass can also be gasified to a mixture of carbon monoxide and hydrogen for catalytic
conversion into methanol. Currently, methanol from biomass is projected to cost about
S0.85/gallon, and with improved technology, a cost of $0.50/gailon could be realized.
Catalytic processing of pyrolytic oils from biomass produces a mixture of olefins that can be
reacted with alcohols to form ethers such as methyl tertiajy butyl ether (MTBE) for use in
reformulated gasoline to reduce emissions. Costs could be competitive today for olefins
from MSW. Because biofuels technologies require little if any fossil fuel inputs, carbon is
recycled through their use, reducing substantially the net amount of carbon dioxide released
to the atmosphere.
6-39
-------
INTRODUCTION
The United States contributes about 25% of the carbon dioxide (C02) released from
fossil fuels to the atmosphere, with transportation fuels accounting for about 27% of that
amount (1). Carbon dioxide is believed to be the most important greenhouse gas, trapping
nearly 50% of the radiation that could lead to global climate change (2), and significant
benefit would be gained by development of fuels that do not contribute to the buildup of
C02 in the atmosphere. Because significant air pollution problems are attributed to gasoline
and other conventional fuels, alternative transportation fuels are of particular interest for
reducing carbon monoxide and smog in our cities.
In addition to concern about the effects of fuel use on the environment, the U.S.
economy depends strongly on unstable sources of imported petroleum. As a result, the
nation experienced several oil price shocks as foreign producers controlled supplies of oil
in the 1970s. In 1990, the price of oil and gasoline increased again because of the Iraqi
invasion of Kuwait, reminding us that oil and natural gas represent the weakest links in the
U.S. energy supply. The United States imports about half of the petroleum it uses annually,
and even though natural gas imports are now less than 6% of usage, gas and oil imports in
the year 2000 are predicted to rise to 11% and 56% of the 23.6 and 37.2 quads of
consumption, respectively. Furthermore, petroleum imports are responsible for about 40%
of the trade deficit of the United States (3). About 97% of the energy consumed by the
transportation sector is derived from petroleum, making this important segment of our
economy particularly vulnerable to disruptions in unstable sources of oil.
Fuels known as biofuels can be produced from many plant materials and waste
products such as agricultural wastes and municipal solid waste (MSW) that together are
called biomass. Several oxygenated biofuels such as ethanol reduce carbon monoxide
emissions when blended with gasoline. A number of biofuels can be substituted directly for
conventional transportation fuels to reduce urban air pollution. Biofuels can also be used
in the residential, industrial, and utility sectors. Substantial improvements have been made
in the technology for producing the liquid and gaseous biofuels that this nation needs most,
and these fuels now have the opportunity to make a strong impact on our fuel use.
This paper focuses on the progress of and prospects for promising biofuels produced
by both biological and thermal reactions. First, the possible contribution of biofuels to our
energy supply will be presented. Then, the technologies for producing biofuels will be
summarized, and some of their important characteristics discussed. The potential benefits
of biofuels for mediating C02 buildup will be illustrated.
BIOFUELS PRODUCTION TECHNOLOGIES
BIOMASS AVAILABILITY AND COST
Plants use the sun's energy to convert C02 and water into simple sugars through
photosynthesis. These sugars can be stored directly in plants such as sugar cane or
combined to form starch for plants such as corn. All plants join sugars together to form the
structural carbohydrate polymers cellulose and hemicellulose, which together with lignin
6-40
-------
support the plant. The cellulose, hemicellulose, and lignin components represent the largest
fractions of plant matter and can be termed cellulosic biomass or just biomass. When we
burn biomass, the stored energy from the sun is released. The energy content of biomass
can also be thermally or biologically transformed to liquid or gaseous fuels that integrate
well with our existing fuel distribution and use infrastructure.
Although the outward appearance of the various forms of cellulosic biomass such as
wood, grasses, municipal solid waste (MSW), and agricultural residues is different, all of
these materials are quite similar in composition. Cellulose is generally the dominant
fraction, representing about 40% to 50% of the material by weight, while the hemicellulose
portion represents 20% to 40% of the material. The remaining fraction is predominately
lignin with lesser amounts of substances called extractives. The cellulose polymer is
composed of glucose sugar while arabinose, mannose, xylose, glucose, and other sugars make
up hemicellulose. Although most forms of cellulosic biomass are low in cost, the historical
costs of conversion to liquid and gaseous biofuels have been too high to allow economic
application of these technologies on a large scale.
It is estimated that about 190 million acres of land could be used to grow energy
crops dedicated for the production of biofuels. For an average productivity of 9
tons/acre/year, about 1.7 billion tons of cellulosic biomass could be harvested each year (4).
If accessible underutilized wood, agricultural residues, and MSW are included, about 2.5
billion dry tons per year of cellulosic biomass could be produced at prices from $18 to $65
per dry ton (1,5,6). As a perspective on the size of this resource, about 250 billion gallons
of liquid fuels could be produced from this quantity of biomass compared to the
approximately 115 billion gallons of gasoline used in the United States.
FUEL ETHANOL
More than 3 billion gallons of ethanol produced from sugar cane is used in Brazil
each year. However, U.S. sugar prices are controlled at about $0.20/pound, a price that is
too high for fuel production. Instead, about 1 billion gallons of ethanol are made from corn
in the United States each year. Ethanol from corn currently sells for about $1.20 to
$1.30/gailon and is competitive with state and federal tax incentives. Ethanol blends from
starch and sugar crops comprised as much as 8% of the U.S. gasoline market in 1987, up
from less than 1% in 1981. There are about 50 U.S. fuel ethanol manufacturing facilities
that use corn and other grains as feedstocks.
Over the years, several processes have been studied for conversion of cellulose-
containing biomass to ethanol catalyzed by dilute acid, concentrated acid, or enzymes known
as cellulases. In each option, the feedstock is pretreated to reduce its size and open up the
structure. The cellulose fraction is hydrolyzed by acids or enzymes to produce glucose sugar,
which is subsequently fermented to ethanol. The soluble xylose sugars derived from
hemicellulose are fermented to ethanol as well, while the lignin fraction can be burned as
fuel to power the rest of the process, converted into octane boosters, or used as a feedstock
for production of chemicals.
6-41
-------
Dilute acid systems typically have low ethanol yields of 50% to 70%. Concentrated
sulfuric or halogen acid options achieve the high yields required, but the acids must be
recovered at a cost substantially lower than the cost at which these inexpensive materials
are produced in the first place, a difficult requirement. Enzyme-catalyzed options provide
the high yields of ethanol necessary for economic viability, under mild conditions, with low
concentrations of enzyme. In addition, enzyme-catalyzed processes have tremendous
potential for technology improvements that could bring the selling price of ethanol down to
levels competitive with those of existing fuels. Enzymes are also biodegradable and environ-
mentally benign. At this time, the simultaneous saccharification and fermentation (SSF)
enzyme-based process has emerged as the favored route to achieve low-cost fuel ethanol
production within a reasonable time frame (7).
In the SSF process, the celiulosic biomass is first pretreated to open up the biomass
structure and facilitate subsequent processing. Several options have been considered for
biomass pretreatment including steam explosion, acid catalyzed steam explosion, ammonia
fiber explosion, and organosolv, but the dilute acid option appears to have the best near-
term economic potential. In this process, about 0.5% sulfuric acid is added to the feedstock,
and the mixture is heated to around 140° to 160° C for 5 to 20 minutes. Under these
conditions, most of the hemicellulose is broken down to form xylose and other sugars,
leaving behind a porous material of primarily cellulose and lignin that is more accessible to
enzymatic attack.
Following pretreatment, a portion of the pretreated biomass is used in an enzyme
production vessel to support growth of a fungus that produces cellulase enzyme. Then, the
cellulase enzyme is added to the bulk of the pretreated substrate along with yeast or other
fermentative microorganisms. The enzymes catalyze the breakdown of the cellulose by the
so-called hydrolysis reaction to form glucose sugar; the yeast or other suitable microbe
ferments the glucose to ethanol. The presence of yeast along with the enzymes minimizes
sugar accumulation in the vessel, and since the glucose produced during breakdown of the
cellulose slows down the action of the cellulase enzymes, higher rates, yields, and
concentrations are possible by consuming the sugar as it is released. Additional benefits are
that this process reduces the number of fermentation vessels to about half that for separate
hydrolysis and fermentation steps and that the presence of ethanol makes the fermentation
mixture less likely to be invaded by unwanted microorganisms. Finally, the ethanol is
separated from the rest of the fermentation broth in a purification step.
The xylose and other sugars released from the hemicellulose polymers are often
predominantly five-carbon sugars that are not as readily converted to ethanol as glucose, and
until recently, these sugars had to be discarded. However, several options have been
developed for using xylose. At this time, the most promising option appears to be the use
of genetically engineered bacteria (8,9) or some new yeast strains for ethanol production.
Because lignin represents a significant fraction of celiulosic biomass, it is important
to derive value from lignin. Lignin has a high energy content and can be used as a boiler
fuel. Generally, the amount of lignin in most feedstocks is more than sufficient to supply
all the hea;t and electricity required for the overall ethanol production process, as well as
to generate excess heat or electricity. Thus, additional revenue can be derived from
6-42
-------
electricity exports from the plant (10). The electricity sold for current plant designs is
equivalent to about 8% to 10% of the energy content of the ethanol product, and greater
revenues are likely as the technology is improved to require less process heat and electricity.
Alternatively, lignin could be converted into chemicals or octane boosters such as methyl
aryl ethers.
Progress on the enzyme-catalyzed processes to convert cellulosic biomass into fuel
ethanol has been substantial over the last ten years, with projected selling prices dropping
from about $3.60/gallon in 1980 (11) to only about $127/gallon (12). This selling price
reduction is due to improvements in enzymes to achieve higher rates, yields, and
concentrations with lower loadings, proper selection of fermentative microbes, and advances
in xylose fermentations through genetic engineering.
Significant opportunities still exist to lower the selling price of ethanol from cellulosic
biomass at the plant gate to $0.67/gallon. Key target areas include improved glucose and
xylose yields from pretreatment, increased ethanol yields to 90% or greater from cellulose
and xylose fermentations, decreased stirring and pretreatment power requirements, better
productivities through continuous processing, low-cost production of octane enhancers or
chemicals from lignin, increased ethanol concentrations, and reduction of fermentation
times. Feedstock costs are a significant fraction of the final product selling price, so
improvements in feedstock production, collection, and genetics could provide additional cost
reductions through economies of scale for larger ethanol plants, decreased feedstock costs,
and less non-fermentable feedstock. Many of these goals have been met individually and
the evidence that the rest can be achieved is great; the primary need is to meet them
simultaneously. Fortunately, there are enough options for lowering the selling price of
ethanol so that not all the technical jgoals must be achieved to reach the target.
BIODIESEL FROM MICROALGAE
Microalgae are single-celled plants that contain photosynthetic machinery driven by
the sun's energy to combine C02 and water to form a variety of products. Algae are
particularly unique in their ability to produce a high fraction of their total weight (about
60% or more) as lipid oils or triglycerides. Lipids are hydrocarbons with a higher energy
density than that of the carbohydrates plants typically produce. Although these algal oils
can be used directly in diesel engines, they can also be readily converted into esters that
more closely match diesel fuel properties and burn more cleanly (13). Diesel fuel currently
supplies about 17% of the energy used for transportation in the United States. Production
of diesel fuel from algae complements ethanol manufacture from cellulosic biomass in that
as ethanol displaces the fraction of petroleum converted into gasoline, substitutes must be
found for the fraction now used to manufacture diesel fuel.
Microalgae grow well over a wide range of temperatures in high-salinity water that
is unsuitable for other purposes. Therefore, it may be possible to use the abundant high-
salinity water in aquifers coupled with inexpensive flat land available in the Desert
Southwest to grow microalgae to produce biodiesel. Shallow uncovered ponds or raceways
could be created to produce algae with slowly rotating paddle wheels used to circulate the
water and provide mixing. Carbon dioxide from power plant flue gas would be injected into
6-43
-------
the ponds to promote algal growth. The rapidly growing microalgae would be harvested,
and the lipid oils would be extracted for conversion into ester fuels (13).
Microalgal ponds are very efficient in their uptake of C02, recovering about 90% of
the gas injected into a pond. Thus, they provide an effective means of COz recovery from
power plants. Of course, when ester ftjels are burned, the C02 captured by the algae is
released to the atmosphere, but almost twice as much energy is produced for a given
amount of C02 released as would be possible without the use of algal ponds. As a result,
a coal-burning power plant coupled to an algal pond would contribute far less C02 to the
atmosphere than a conventional gas-fired plant on a total energy released basis (14). Of
course, similar benefits could be provided to a gas-fired power plant. If the lipid oils were
converted to a chemical for production of durable goods, the carbon could be sequestered
for a longer term, reducing the impact of fossil fuel use on C02 accumulation. Use of
biodiesel also has a low sulfur impact on the environment.
Progress on technology for producing oil from microalgae has been considerable. A
number of strains have been collected that are tolerant to high salinity, high light intensity,
wide temperature variations, and extreme temperatures. Many of these strains grow rapidly
and produce about 60% of their weight as lipids when they are deprived of key nutrients
such as silicon for diatoms or nitrogen for green algae (15). The enzyme acetyl Co-A
carboxylase (ACC) has been identified as a key catalyst in lipid oil synthesis (15,16), and
research is now focused on developing techniques to genetically enhance lipid oil synthesis
by controlling the genes responsible for ACC production.
The projected price of biodiesel production from algae has dropped from
approximately $18/gallon in the early 1980s to around $3.50/gallon now. Opportunities
have been identified to reduce the price to about $1.00/gallon. The primary need is to
enhance the growth rate of algae while achieving high lipid oil concentrations. However,
these estimates are based on commercial prices for C02, which represents a major share of
the oil production cost. If fines were levied for C02 releases into the atmosphere, C02
prices would drop drastically or users might be paid to accept this gas, and algal oil
production could be cost competitive much sooner.
BIOGAS
Natural gas, which is primarily methane, is considered an environmentally clean and
economically attractive fuel with a high energy content. It is widely used for industrial,
residential, and utility applications. It also has potential merit for transportation markets,
but it is nonrenewable. Biogas, a mixture of approximately equal portions of methane and
COz, is a medium-Btu gas formed by a biological process called anaerobic digestion, in
which complex organic compounds are decomposed by microorganisms. Anaerobic digestion
has been used in the United States since early in the twentieth century to stabilize and
reduce the volume of municipal sludge before disposal. In typical installations, the biogas
produced in the process was flared rather than recovered as an energy source. Anaerobic
digestion is also used to remove soluble wastes from chemical plant effluents. However,
although the anaerobic digestion processes now operating are effective in meeting waste dis-
posal requirements, current systems were not designed for cost-effective energy production
6-44
-------
and are not well suited to utilization of solid cellulosic biomass. In sanitary landfills,
naturally occurring anaerobic bacteria break down the biodegradable fraction of the MSW
buried there to form biogas although the gas production rates and yields vary widely. Only
a small percentage of that gas is now being economically recovered.
In the anaerobic digestion process, one group of bacteria breaks down cellulose and
other complex molecules enzymatically into simple sugars and other monomers. Then, other
types of bacteria digest these products, producing organic acids that are in turn broken down
to form still smaller molecules of acetate, formate, hydrogen, and COz. Finally, specialized
bacteria, called methanogens, use these compounds to produce methane and C02. When
C02 is removed from biogas, the methane-rich product is a high-Btu gas that can be directly
substituted for natural gas. Thus, this biological pathway directly converts organic matter
into a gas that can be used directly in a boiler or processed to be compatible with the
existing distribution infrastructure.
If MSW is employed as the substrate, anaerobic digestion also provides an
environmentally sound disposal method. In the anaerobic digestion of MSW, the solid waste
is shredded and ferrous materials are removed. Generally, it is also necessaiy to separate
the extra-fine and oversized materials for landfill disposal. The mixture is fed into digesters,
and the microbial process converts about half the solid waste into a gas composed of about
equal fractions of methane and C02. The solids left after digestion is complete are
dewatered for disposal. For a tipping fee of $20/ton for the MSW processed, a methane
selling price of about $4.50/MBtu (106 Btu) is estimated for established technology.
Several improvements have been made in the anaerobic digestion process. Stratified
operation of the digester can result in higher solids concentrations and achieve 10% higher
methane yield than a conventional vessel. Technology has also been demonstrated to enrich
the methane content of the product gas from an anaerobic digester to near pipeline quality
by recirculating a leachate stream from the digester through an air stripping unit or other
C02 desorption process. Novel anaerobic digestion units have been operated at 35% to
40% solids concentrations with gas generation rates of 7 to 9 times those possible with
conventional devices, which are limited to 6% to 10% solids levels (17). Such devices
reduce the volume of the containment vessel and thus decrease vessel costs per amount of
methane generated.
Continued research could result in further reductions in methane cost to
approximately $2.00/MBtu (18). Research is needed to identify the organisms present in
anaerobic digesters and clarify their complex interactions. Specific organisms can then be
selected for genetic breeding and manipulation to allow operation of anaerobic digesters at
optimal conditions. Feedback and control mechanisms should be developed to maintain
stable operation of digesters (19). Engineering efforts should be undertaken to develop
large-scale reactor designs that can process high concentrations of solids to decrease the cost
of biogas production. Work is also needed on landfill gas recovery to understand the effect
of atmospheric conditions on gas flow, evaluate microbial populations that produce landfill
gas, and improve gas generation and capture methods.
4.
6-45
-------
METHANOL FROM BIOMASS
Biomass can be thermally gasified to produce synthesis gas (syngas) rich in carbon
monoxide and hydrogen; syngas can then be catalytically converted to methanol. Methanol
production includes steps for feed preparation, thermal gasification, methanol synthesis, and
gas conditioning and cleanup. Feed preparation typically employs well developed equipment
for biomass drying, size reduction, and feeding. Methanol synthesis from carbon monoxide
and hydrogen is a well established technology. In order to achieve commercial application
of methanol production from biomass, low-cost technology is needed for thermally gasifying
the biomass to carbon monoxide and hydrogen and preparing the gas stream for catalytic
formation of methanol.
The gasification step can be carried out by controlled direct addition of air or oxygen
to the gasifier to produce heat and drive the breakdown of biomass to form carbon
monoxide and hydrogen. In such processes, part of the biomass is burned to provide the
heat required to drive the formation of the target gases. Alternatively, indirectly heated
gasifiers rely on transfer of heat from an external source through a heat exchange device to
break down biomass (20,21). Indirect gasifiers may have some cost advantages relative to
units that are directly fed air or oxygen. However, further engineering evaluations and
economic studies are needed to establish the relative merits of the alternative processes.
Although the raw syngas is rich in carbon monoxide and hydrogen, it cannot be
directly processed in the catalytic synthesis unit because it contains significant amounts of
impurities including particulate matter, methane, tar, and various light hydrocarbons. In
addition, the ratio of carbon monoxide to hydrogen must be adjusted to that required for
methanol synthesis. Research has shown that the gas can be cleaned up and the
composition adjusted for methanol synthesis. However, further work is required to
determine the fate of organic impurities and methane; establish the effect of hydrogen,
carbon monoxide, and C02 on tar and methane removal; estimate useful catalyst life; and
demonstrate the ability of catalysts to destroy tars and reduce methane under realistic
operating conditions with various feedstocks.
Currently, methanol is estimated to cost about $0.85/gallon from biomass feedstocks
costing about $42/ton. Several opportunities have been identified that would make
methanol from biomass competitive with gasoline as a neat fuel. Direct syngas conditioning
in one step would replace expensive quenching and scrubbing operations to remove tars and
subsequent reheating and steam reforming to reduce excessive levels of methane. We need
to develop catalysts that meet lifetime requirements, as well as better gas cleanup systems.
Concepts must also be tested at a reasonable scale to establish operational parameters and
commercial potential. Improvements in feedstock costs, as discussed for ethanol production,
would also benefit methanol economics. Successful demonstration of these improvements
could drop the cost of methanol to about S0.50/gallon, a price competitive with gasoline
from oil at $25/barrel.
6-46
-------
REFORMULATED GASOLINE COMPONENTS (RGCs)
Concerns about urban air pollution have led to enactment of legislation such as the
Clean Air Act Amendments of 1990, which require measures be taken to improve air
quality. Use of neat fuels such as ethanol and methanol are one way to reduce emissions
of unburned hydrocarbons and carbon monoxide, which contribute to smog formation. In
addition, gasoline compositions are being changed or "reformulated" to reduce their
contribution to air pollution. Oxygenates such as ethanol and ethers are added to gasoline
to improve fuel combustion and reduce the release of smog-forming compounds as well as
carbon monoxide. The Clean Air Act Amendments of 1990 mandate the use of 2.7%
oxygen in fuels for several ozone and carbon monoxide non-attainment regions, thus causing
a large demand for oxygenates such as ethanol and methyl tertiary butyl ether (MTBE).
Ethers such as MTBE and ethyl tertiary butyl ether (ETBE) are made by reacting the
appropriate alcohol (i.e., methanol or ethanol, respectively) with isobutylene. Currently,
isobutylene is derived from fossil sources, but these compounds can also be made by fast
pyrolysis of biomass followed by catalytic reaction of the pyrolysis products to form olefins.
Tlie olefins in turn can be reacted with alcohols to form RGCs such as ETBE and MTBE.
Fast pyrolysis rapidly heats the biomass to temperatures at which the predominant
pyrolysis reactions form oxygenated crude oil vapors, rather than char, water, or gases. A
vortex reactor forces biomass particles to slide along the externally heated reactor wall, and
the close contact between the particles and the wall produces high rates of heat transfer.
As a result, the surface of the biomass particle is pyrolyzed and removed, but the bulk of
the particle is still unheated through an ablative pyrolysis phenomena (22£3).
The oxygenated crude pyrolysis oil vapors produced in the vortex reactor are
converted into olefins and other products through a thermal cracking operation. Extensive
studies of these reactions have typically shown that some olefins and other hydrocarbons of
interest result but in yields that are not high enough to be economically interesting.
However, zeolite catalysts such as HZSM-5 used to convert methanol to gasoline crack the
pyrolysis oil vapors with dramatically higher olefins production. Researchers have identified
operating conditions with this zeolite catalyst in a slipstream reactor, and low coke yields
and relatively high yields of high-octane alkylated aromatics and gaseous olefins were
achieved. These products give a very high octane blending stock for use with reformulated
gasoline (23,24).
The offgas from the catalytic cracking reactor contains a considerable amount of
gaseous olefins. Catalytic conversion of these olefins to larger hydrocarbon molecules in a
secondary catalytic reactor has been studied to facilitate condensation and recovery of these
materials. A compressor pressurizes the gases that enter the secondary catalytic reactor.
Primarily isoparaffins boiling in the gasoline range have been generated from the gaseous
olefins. Changes in the catalysts, coreactants, temperatures, and pressures selected could
change the products from the secondary reactor.
Preliminary evaluations of the economics of the combined pyrolysis and catalytic
conversion process are encouraging, especially when refuse-derived fuel (RDF), a low-cost
feedstock derived from MSW that gives enhanced yields of olefins compared to those from
6-47
-------
wood, is used as the feedstock. Research on novel catalysts will increase both the olefin
yields and the selectivity toward olefins versus aromatics. Initial catalyst screening in the
laboratory has already identified some promising candidates, and the results suggest that
yields could be improved substantially.
Preliminary process evaluations of the advanced catalytic process suggest that mixed
ethers could be produced for $0.67/gal while they would cost about $0.96/gal based on
existing technology (25). However, the vortex reactor represents a new technology, which
carries a considerable risk because of the lack of current industrial experience. Therefore,
further research is required in areas such as selectivity and catalyst life. In addition, the
ability to scale up the fast pyrolysis of biomass to produce condensable pyrolysis vapors must
be demonstrated.
IMPACT OF BIOFUELS ON CARBON DIOXIDE ACCUMULATION
In this section, the impact of using biofuels on C02 accumulation and the potential
for global climate change will be discussed. An example will be presented based on ethanol
production from cellulosic biomass, but the concepts are applicable to any of the biofuels
that have been presented here.
Carbon dioxide released during fermentation and ethanol combustion is recycled back
to grow new biomass, replenishing that harvested for ethanol manufacture. In addition, the
lignin in the feedstock can be burned to provide all the process heat and electricity needed
to drive the conversion process, and some heat or electricity is left to sell for added revenue.
Thus, provided new trees or other biomass are planted to replace those that are harvested
for energy, C02 provides the key link between the fuel and the biomass resource, and C02
does not accumulate in the atmosphere. Fossil fuels used in the production and transport
of biomass and ethanol are the only sources that can lead to C02 accumulation.
Considerable controversy and confusion exists about the amount of fossil energy
required to produce ethanol and the interpretation of this information for energy efficiency
and the impact on potential global climate change. Energy inputs must be properly
accounted for and the performance of ethanol must be properly compared to that for fossil
fuels if comparisons are to be made between fuels.
ENERGY USE
Figure 1 summarizes energy flows for the production of ethanol from cellulosic
biomass (1). Modest energy inputs are required to produce cellulosic biomass because
cultivation and fertilizer needs are not large. In addition, no process energy input is shown
because the lignin contained in the feedstock can be used as a boiler fuel, and the amount
of energy contained in the lignin is sufficient to produce process heat and electricity for the
overall process. In addition, the equivalent of 8% of the ethanol fuel value can be sold as
electricity (9), thereby displacing about 3 times that amount of fossil energy inputs for
electricity production by power companies.
6-48
-------
Electricity
E « 0.08'
Agricultural inputs
A-0.15'
Raw material transport
T - 0.04"
Feedstock
Production
Processing
-f Utilities
Energy crops
PretreatiTiem
Biological steps
• Celiulase production
• Hydrolysis
• Fermentation
Ethanol recovery
-( Residue processing
Chemical inputs
C - 0.01
Distributor)
0 - 0.01
Fuel
ethanol
F - 1.0
Plant
' amortization
P - 0.04
Process effluents
Figure 1. Energy requirements and outputs for production of ethanol from cellulosic
biomass based on one unit of ethanol energy output
If we add the total amount of fossil energy inputs, about 19,000 Btu/gallon of ethanol
are required for ethanol production from cellulosic biomass. Alternatively, 3 times the
electricity produced could be subtracted from the fossil fuel requirements to account for the
fossil fuel that would be used in any event for producing that quantity of electricity (26).
In this case, the net amount of fossil fuel use, as shown in Table 1, is only about 800
Btu/gallon of ethanol. For comparison, production of gasoline requires about 14,000
Btu/gallon of gasoline produced. Gasoline has a lower heating value of about 115,000
Btu/gallon, compared to ethanol at 76,000 Btu/gallon.
CARBON DIOXIDE RELEASED
An estimate of the contribution of ethanol production to C02 accumulation in the
atmosphere can be performed by weighing the quantities of fossil fuels used according to
the amount of COz released by each. For the purposes of this discussion and with reference
to Figure 1, natural gas is assumed to be the fuel source for agricultural (A), chemical (C),
and plant amortization (P) inputs; petroleum is assumed for transportation (T) and
distribution (D). Combining COz release data for these sources with the energy
requirements presented in Figure 1 gives the results presented in Tabie 1 for ethanol
derived from cellulosic biomass. Only C02 is included that is produced by combustion of
fossil fuels since C02 generated during fermentation of biomass cellulose and hemicellulose
and combustion of lignin and ethanol can be recycled to grow new biomass to replace that
harvested for energy production.
It is interesting to note that if the fossil equivalent for production of electricity is
subtracted from the fossil fuel sources for ethanol production from cellulosic biomass, a
negative C02 contribution results (Table 1). This outcome is due to the low usage of fossil
fuels for conversion of cellulosic biomass to ethanol and the displacement of electricity in
6-49
-------
t
TABLE I. CARBON DIOXIDE RELEASED IN ETHANOL PRODUCTION FROM CELLULOSIC BIOMASS
Process
Feedstock
Production
Raw
Material
Transport
Chemical
Inputs
Process
Energy
Plant
Amortiza-
tion
Fuel
Distribu-
tion
Total
Fossil C02
Coproduct
Displacement
of CO,
Net Fossil
CO,
Assumed
Fossil Fuel
Btu/Gallon
Ethanol
Carbon
Dioxide
Released
lb/gallon
Natural Gas Petroleum
11,400 3,000
1.5
0.5
Natural
Gas
800
0.1
None
Natural Gas Petroleum
3,000 800
0.4
0.1
19,000
2.6
(18,200)
(4.6)
800
(2.0)
Note: Calculated on the basis of the exported electricity displacing coal-generated electricity, which was generated at 33% efficiency
cn
cn
o
-------
the grid. To produce an equivalent amount of electricity from fossil resources such as coal
would actually produce more C02 than the total released during ethanol production, and
subtracting the amount of C02 that would have been released anyway for generation of that
quantity of electricity gives a net credit of 2 pounds of COj/gallon of ethanol produced.
From the data in Table 1, it appears that production of ethanol from cellulosic
biomass would be a minor contributor of C02 to the atmosphere. This is because all of the
process heat is produced by combustion of lignin, a renewable feedstock. In addition, few
fossil fuel inputs are needed to produce cellulosic biomass. Biomass sources of fuels and
fertilizers could be substituted for the fossil fuels assumed in this analysis, resulting in no
net C02 release for ethanol from cellulosic biomass. By way of comparison, gasoline use
releases a total of 190 pounds of COz/MBtu or 21.8 pounds/gallon of gasoline.
FUEL UTILIZATION
In the United States, ethanol is currently blended with gasoline at 10%
concentrations. As mentioned earlier, the energy content of ethanol is 76,000 Btu/gallon;
gasoline contains about 50% more energy at 115,000 Btu/gallon. By accounting for each
of these components, the energy content of the blend is about 111,000 Btu/gallon. If we
assume that the range of a vehicle is proportional to the energy density of the fuel, then
1.036 gallons of blend would be required to travel the same distance as one gallon of
gasoline. On the other hand, data from Southwest Research Institute have shown that
there is no statistically significant difference in the mileage for a 10% blend versus that of
regular gasoline. In effect, this evidence suggests that the ethanol blended with gasoline has
an energy density equivalent to 115,000 Btu/gallon.
Use of neat ethanol can also be considered in at least two different ways. First, the
amount of one fuel needed to travel the same distance as for another fuel can be
determined by the ratio of the lower heating values of the two fuels. Thus, about 50% more
ethanol would be required than for gasoline to give the same service. For an engine
designed for gasoline use with only modifications in timing and ainfuel ratio to allow
combustion of ethanol, ethanol fuel would give such a range. However, because ethanol has
more favorable fuel properties such as a higher octane and heat of vaporization than
gasoline, an engine optimized for ethanol can be 20% to 30% more efficient than a gasoline
engine (1,6). Assuming the latter results in a ethanol driving range of about 80% of that
of gasoline.
COMPARISONS OF FOSSIL FUEL USE AND CARBON DIOXIDE RELEASE
Table 2 presents a comparison of the fossil fuel requirements for ethanol from
cellulosic biomass and gasoline. Consideration is given to both blends and neat fuel use and
to comparison of the amount of ethanol required based on lower heating value and
performance. The lower end of the range shown is with credit given for fossil fuel
displacement for electricity production, while the higher end does not account for excess
electricity production. Blends of ethanol from cellulosic biomass with gasoline always yield
lower fossil fuel use than gasoline. Furthermore, for neat fuel markets, ethanol production
from cellulosic biomass requires one-fifth or less fossil fuel input than gasoline, depending
6-51
-------
TABLE 2. FOSSIL FUEL USE AND C02 RELEASED FROM CELLULOSIC
BIOMASS
Form of Fuel Use Fossil Fuel Used
(Btu/gal of gasoline
equivalent)
COz Released
(lb COj/gal of gasoline
equivalent)
Blends Based on Lower Heating Value
Ethanol
120,400 - 122,200
20.1 - 20.6
Blends at Same Range as Gasoline
Ethanol
116,200 - 118,000
19.4 - 19.9
Neat Fuel Based on Lower Heating Value
Ethanol
1,200 - 28,700
-3.0 - 3.9
Neat Fuel Based on Improved Efficiency
Ethanol
Gasoline
1,000 - 23,800
129,000
-2.5 - 3.2
21.8
on the accounting given for electricity production and the assumed efficiency of ethanol
utilization.
Also presented in Table 2 is the amount of C02 released when a vehicle is propelled
the same distance by ethanol as gasoline. Again, based on the low use of fossil fuels in
production of ethanol from cellulosic biomass, this fuel scores very well in minimizing C02
emissions that could contribute to global climate change. If we subtract the C02 emissions
that would have resulted from fossil fuels to generate the amount of electrical, energy
produced in the ethanol plant, the net effect is that ethanol removes C02. It may be more
appropriate to assign the C02 released to both electricity and ethanol based on the relative
energy contributions and compare each to the alternative, but the benefits would still be
substantial.
A wide range of fuel products, which can reduce our vulnerability to disruptions in
fuel supplies and improve our balance of trade deficit, can be produced from biomass, a
renewable, abundant feedstock. These products include ethanol, methanol, methane gas,
biodiesel, and olefins for production of RGCs. All of these biofuels can be used in the
transportation sector, some as gasoline additives to reduce emissions of carbon monoxide
and smog-forming compounds and others as substitutes for fossil fuels. Some can also be
employed for residential, utility, and industrial applications. Substantial progress has been
made in reducing the price of biofuels production, and goals have been defined to further
reduce that cost. Many of these fuels are now ready for introduction into our energy sector
CONCLUSIONS
6-52
-------
for selected applications, and their use would improve our environment significantly from
a viewpoint of both local air pollution and the potential for global climate change. Because
only modest amounts of fossil fuels would typically be used to produce biofuels, the net
release of C02 to the atmosphere would be small. Furthermore, for celiulosic biomass, all
of the energy inputs in the overall process could ultimately be derived from renewable
feedstocks, thereby avoiding any net release of C02.
The work reported is funded by the Biofuels Systems Division of the U.S.
Department of Energy. The work described in this paper was not funded by the U.S.
Environmental Protection Agency and therefore the contents do not necessarily reflect the
views of the Agency and no official endorsement should be inferred.
REFERENCES
1. Lynd, L.R., Cushman, J.H., Nichols, RJ., and Wyman, C.E. Fuel ethanol from
celiulosic biomass. Science. 251:1318, 1991.
2. Intergovernmental Panel on Climate Change. J.T. Houghton, GJ. Jenkins, and
J.J. Ephraums, eds. Climate Change-the IPCC Scientific Assessment, Cambridge
University Press, Cambridge, United Kingdom, 1990.
3. Wald, M.I. Greater reliance on foreign oil feared as U.S. output tumbles. New York
Tunes. 18 January 1990.
4. National Renewable Energy Laboratory (Interlaboratory Report). The Potential of
Renewable Energy: An Interlaboratory White Paper, SERI/TP-260-3674, National
Renewable Energy Laboratory, Golden, Colorado, 1990.
5. Wyman, C.E., and Hinman, N.D. Ethanol: fundamentals of production from
renewable feedstocks and use as a transportation fuel. AppL Biochem. BiotechnoL
24/25:735, 1990.
6. Lynd, L.R. Production of ethanol from lignocellulose using thermophilic bacteria:
critical evaluation and review. Adv. in Biochem. Eng./Biotechnol. 38:1, 1989.
7. Wright, J.D., Wyman, C.E., and Grohmann, K. Simultaneous saccharification and
fermentation of lignocellulose: process evaluation. AppL Biochem. Biotechnol. 18:75,
1988.
8. Ingram, I.O., Conway, T., Clark, D.P., Sewell, G.W., and Preston, J.F. Genetic
engineering of ethanol production in Escherichia coli. AppL Environ. MicrobioL
53(10):2420, 1987.
9. Ingram, I.O., and Conway, T. Expression of different levels of ethanologenic
enzymes from Zynomonas mobilis in recombinant strains of Escherichia colL AppL
Environ. MicrobioL 54(2) :397, 1988.
10. Hinman, N.D., Schell, DJ., Riley, CJ., Bergeron, P.W., Walter, PJ. Preliminary
estimate of the cost of ethanol production for SSF technology. AppL Biochem.
BiotechnoL 34/35:639 (1992).
11. Wright, J.D. Ethanol from lignocellulose: an overview. Ener. Prog. 8(2):71, 1988.
12. Chem Systems. Technical and Economic Evaluation: Wood to Ethanol Process, Chem
Systems, Tarrytown, New York, prepared for the Solar Energy Research Institute,
August 1990.
6-53
-------
13. Neenan, B., Feinberg, D., Hill, A., Mcintosh, R-, Terry, K. Fuels from Microalgae:
Technology Status, Potential, and Research Requirements. Solar Energy Research
Institute, Golden, Colorado, 1986.
14. Chelf, P., Brown, L., and Wyman, C. Aquatic biomass resources and carbon dioxide
trapping. Reprint from Proceedings of Electric Power Research Institute conference
entitled "1990 Conference on Biomass for Utility Applications," Oct. 23-25, 1991,
Tampa, FL.
15. Roessler, P. Changes in the activities of various lipid and carbohydrate biosynthetic
enzymes in the diatom Cyclotella cryptica in response to silicon deficiency. Arch.
Biochem. Biophys. 267:521, 1988.
16. Roessler, P. Environmental control of glycerolipid metabolism in microalgae:
commercial implications and future research directions. J. PhycoL 26:393, 1990.
17. Rivard, C., Himmel, M., Vinzant, T., Adney, W., Wyman, C., Grohmann, K.
Development of a novel laboratory scale high solids reactor for anaerobic Digestion
of processed municipal solid wastes for the production of methane. AppL Biochem.
BiotechnoL 20/21:461, 1989.
18. Legrand, R., and Todd, T. Systems Analysis of Municipal Solid Waste Biogasification.
Prepared for the Solar Energy Research Institute, Reynolds, Smith, and Hills, Inc.,
Jacksonville, Florida, 1990.
19. Chynoweth, D., Fannin, K., Jerger, D., Srivastava, V., Biljetina, R. Anaerobic
Digestion of Biomass: Status Summary and R&D Needs/1983. Prepared for the Gas
Research Institute, Chicago, Illinois, Institute of Gas Technology IIT Center, 1984.
20. Chem Systems. Assessment of Cost of Production of Methanol from Biomass, report
DOE/PE-0097P, Chem Systems, Tarrytown, New York, December 1990.
21. Bain, R. Methanol from biomass: assessment of production costs. Presented at the
November 1989 Hawaii Natural Energy Institute Renewable Transportation
Alternatives Workshop, Honolulu, Hawaii, 1991.
22. Diebold, J., and Scahill, J. Biomass to gasoline. Upgrading pyrolysis vapors to
aromatic gasoline with zeolite catalysis at atmospheric pressure. Pyrolysis Oils from
Biomass: Producing, Analyzing, and Upgrading, EJ. Soltes and T-A- Milne, eds.,
American Chemical Society (ACS) Symposium Series 376, American Chemical
Society, Washington, DC, 264-76, 1988.
23. Diebold, J., and Power, A. Data from International Energy Agency sponsored
"Research in Thermochemical Biomass Conversion," 1988.
24. Evans, R., and Milne, T. Molecular beam, mass-spectrometric studies of wood vapor
and model compounds over an HZSM-5 catalyst Pyrolysis Oils from Biomass:
Producing, Analyzing, and Upgrading, EJ. Soltes and T.A. Milne, eds., American
Chemical Society (ACS) Symposium Series 376, American Chemical Society,
Washington, DC, 311-27, 1988.
25. Overend, R. Thermochemical conversion technology. In Proceedings of the Annual
Automotive Technology Development Contractor's Coordination Meeting 1991. Society
of Automotive Engineers, Inc., Warrendale, Pennsylvania, 779, 1991.
26. Ho, S.P. Global impact of ethanol versus gasoline. Presented at the 1989 National
Conference on Clean Air Issues and America's Motor Fuel Business, Washington,
DC, 1989.
6-54
-------
ropRomjcnoN of methanol and power
6E
BY: William Weber
Electric Power Research Institute
Palo Alto, CA. 94303
Ardert B. Walters
Florida Power & Light Company
West Palm Beach, FL. 33407
Samuel S. Tam
Bechtel Group, Inc.
San Francisco, CA 94119
ABSTRACT
Integrated Coal Gasification Combined Cycle (IGCC) is one of the emerging
technologies for electric power generation from coal. Under contract with
Florida Power
-------
6E
INTRODUCTION
The Research & Development Department of Bechtel Group Inc. is
conducting a site-specific study for the coproduction of electric power and
clean by-product fuel in an IGCC plant. This study is being performed for the
Florida Power & Light (FPL) Company and the Electric Power Research
Institute (EPRI).
The major objectives of this study are to first design a base case IGCC plant
that is representative of a base-loaded power plant, and then to identify and to
quantify the potential benefits of a spare gasifier, fuel-grade methanol, and
liquid hydrogen coproduction as additions to the base case plant. This is
significantly different than previous methanol coproduction IGCC studies
which have assumed cyclic operation of the power block with undersized coal
gasification capacity (1).
FPL's Martin Site, located near the eastern shore of Lake Okeechobee, Florida
provides the site data for this study. The major components of the proposed
IGCC plant are: slurry-feed (DESTEC) gasifiers, GE advanced gas turbine
(Model MS7001F), methanol plant based on Liquid Phased Methanol
(LPMeOH) process.
The performance, process design, and cost data for the slurry-feed gasifier
were prepared by DESTEC. GE's Turbine Technology Department estimated
the performance of the latest GE MS7001F gas turbine. This latest gas turbine
unit, fully-loaded with syngas, is capable of generating 192 MW power at an
ambient temperature of 45°F. Chem Systems prepared the design package for
the once-through methanol (OTM) plant which features a slurry reactor
design (LPMeOH). Detail description of the LPMeOH process is available in
various publications (1, 2, & 3).
The results for the power and methanol coproduction are presented in this
paper while the results on the liquid hydrogen coproduction are reported in a
separate paper (4).
6-56
-------
SCREENING STUDY
A screening study was first conducted to define the study cases that would best
illustrate the potential benefits of a spare gasifier and methanol coproduction.
Some of the key factors considered were: plant equivalent availability,
maximum utilization of the spare gasifier, methanol plant size, usage of
methanol produced, and overall plant performance. Two of the cases
developed are summarized here to illustrate and to quantify the benefits of
coproducing methanol and power: (1) a base-loaded IGCC plant without a
spare gasifier as the base case, and (2) a base-loaded IGCC plant with a spare
gasifier and an OTM plant. The block flow diagrams for these two cases are
shown in Figures 1 and 2.
These cases are chosen in order to achieve the maximum plant equivalent
availability with an IGCC plant. As shown in Figure 3, the plant equivalent
availability of a base-loaded IGCC plant is determined to be 86%. A spare
gasification train increases the IGCC plant equivalent availability to 90%. The
gasification plant scheduled and forced outage rates are higher than those of
the gas turbine cycle. If sufficient amount of methanol is produced for backup
fuel, the power block can be operated on methanol when one or more
gasification trains are not available. The resulting plant equivalent
availabilitv will be similar to that in a natural gas-based combined cycle plant
(93%).
PROCESS DESCRIPTION
The base case IGCC plant consists of four slurry-feed gasifiers generating
sufficient medium-Btu syngas to fully load four GE gas turbines at 95°F
ambient temperature. The design coal rate is 7412 short tons per day (stpd).
The oxygen purity is 95% to take advantage of lower power requirement in
the oxygen plant compared to that required to produce 98+% oxygen purity.
The IGCC/OTM plant includes a spare gasifier and an OTM plant. Since the
spare gasifier is also operated at its design capacity, the design coal feed rate is
increased to 9296 stpd. The acid gas removal unit is revised in order to meet
the more stringent sulfur requirement of the catalyst in the methanol plant.
The design of the OTM plant is based on the LPMeOH process. The OTM
plant is sized so that the gasifiers can be operated year-round at the design
capacity. Daily fluctuation of syngas demand from the power block is met by
varying the amount of syngas for duct firing in the HRSG.
6-57
-------
About two-thirds of the syngas from the Coal Gasification Unit is diverted to
the OTM plant. About 30% of the combined carbon monoxide and hydrogen
in the syngas is converted to methanol in the LPMeOH slurry reactor. The
unconverted syngas is expanded to recover some of the energy before it is
combined with the remaining syngas in the Water Saturation Unit feeding
the gas turbines.
RESULTS AND DISCUSSION
PLANT PERFORMANCE
The performance of the IGCC base case piant and the IGCC/OTM plant at an
ambient temperature of 75°F is shown in Table 1. The net power outputs
from the IGCC and the IGCC/OTM plant are 927 and 994 MW, respectively.
The heat rate for the IGCC piant is 8832 Btu/Kwh. The heat rate for the
IGCC/OTM coproduction plant is estimated to be 9125 BTU/Kwh. However,
heat rate is not a good indicator for the thermal efficiency in power
generation from the IGCC/OTM plant because it is highly integrated with the
associated OTM plant. Duct firing in the power block further reduces the
meaningfulness of using the heat rate as the indicator for the IGCC
coproduction plant performance.
The OTM plant produces 978 stpd (100 million gallons per year) methanol.
The OTM plant consumes approximately 50% of the syngas output from a
gasifier. As shown in the latter part of this paper, sufficient amount of
methanol can be produced as backup fuel for the power block.
COST OF ELECTRICITY
As shown in Table 2f the plant investments for the IGCC base case and the
IGCC/OTM plant are estimated to be $1440 and $1592 per Kw, respectively.
The plant investment includes the costs of startup, inventory, owner's costs,
and AFUDC. As shown, the methanol plant cost is about three percent of the
total plant investment.
EPRI's economic model for an investor-owned utility (IOU) scenario was
used to determine the cost of electricity (COE) for the IGCC base case and the
IGCC/OTM plant. The key financial assumptions are shown in Table 3. The
6-P8
-------
COEs for the IGCC base case and the IGCC/OTM plant are 37.0 and 34.6 mills
per Kwh, respectively. As expected from its relatively low capital investment
cost for the OTM plant, methanol coproduction lowers the COE of the IGCC
plant.
As shown in Figure 4, methanol coproduction does not have any negative
impact on the COE of the IGCC plant if the methanol price is at S0.35 per
gallon escalated merely at the inflation rate of 6%.
BENEFITS OF METHANOL COPRODUCTION
The immediate benefits of the methanol coproduction scheme to the IGCC
plant can be illustrated in two scenarios. In the first scenario, when a gasifier
is down unexpectedly, syngas can be diverted to the power block so that power
generation is not interrupted. In this case, the backup fuel is the syngas.
Thus, overall plant thermal efficiency remains the same as the base IGCC
plant and is not decreased as would be the case if methanol or another form
of "stored energy" were employed as backup fuel.
The second scenario involves the planned shutdown of a gasification train.
Because the gasification train has to be shut down for a longer period of time
than the combined cycle plant for planned maintenance, stored methanol can
be used in the power block when syngas is not available. The combined effect
of these two scenarios is to raise the plant equivalent availability of the power
block to 93%, the same availability as the natural gas-based combined cycle
plant. The amount of backup methanol fuel is estimated to be about 24
million gallons per year.
The methanol coproduction concept can also reduce the potential technical
risk of the gasifier, a relatively new technology for power generation industry.
The gasifier risk abatement can be best illustrated by examining the effect of
gasifier component availability on the annual methanol production rate and
the overall IGCC plant equivalent availability. The gasifier component
availability is the probability that the gasifier is in operating condition at any
given time. In this study, the design value for the gasifier component
availability is 96%.
As shown in Figure 4, the IGCC plant equivalent availability is directly
proportional to the gasifier component availability. The IGCC /OTM plant
equivalent availability is 90% when the gasifier component availability is
96%. In the unlikely event, the gasifier component availability should be
84%, the IGCC plant equivalent availability should be 86%. If 61 million
gallons of methanol is available annually as backup fuel, the IGCC/OTM can
6-59
-------
be operated at the same equivalent availability (93%) as that of a natural gas-
based combined cycle plant.
CONCLUSIONS
An IGCC/OTM coproduction plant design has been developed according to
actual FPL dispatched utility operation. With the most pessimistic methanol
price forecast, results of this study indicate that methanol coproduction can
offset the cost of a spare gasifier and increase the new power output of an
IGCC plant with the same number of working gasifiers and gas turbines.
The proposed IGCC/OTM plant configuration can increase the IGCC plant
equivalent availability from 86% to 93%, the equivalent availability of a
natural gas-based combined cycle plant.
The methanol produced can also help to reduce the risk of lower than
expected gasifier availability. If the gasifier component availability were
decrease from 96% (the design basis) to as low as 84%. sufficient methanol
would be available to maintain the combine cycle plant operation at its 93%
plant equivalent availability.
The work described in this paper was not funded by the U.S. Environmental
Protection Agency and therefore the contents do not necessarily reflect the
views of the Agency and no official endorsement should be inferred.
6-60"
-------
Table 1
IGCC Plant Performance Summary (Slurry-Feed Gasifier)
IGCC Base IGCC with
Case the OTM plant
Ambient Temperature. °F 75 75
Number of Gasifiers (Total/Spare) 4/0 5/1
Number of GT/ST/HRSG 4/2/2 4/2/2
Coal Feed Rate, stpd 7412 9296
Products:
Fuel-grade methanol, stpd 0 978
(as 100% methanol)
Gross Power, MW 1035 1158
Net Power Output, MW 927 994
Heat rate, Btu/Kwh (HHV) 8832 9125 (Note)
Note: The estimated heat rate only includes the equivalent
amount of syngas for electrical power generation
6-61
-------
Table 2
IGCC Plant Cost Summary (Slurry-Feed Gasifier)
(1st Qt. 1990 $/Kw)
IGCC IGCC with OTM
Base Case Methanol Plant
Net Power Output, MW 926.7 993.6
Plant Units
Coal Gasification Unit 295 320 (20 %)
Air Separation 144 188 (12 %)
Power Block 422 449 (28 %)
Methanol Plant 0 52 (3 %)
General Facilities 214 178 (11 %)
Startup, Inventory, Owner's Cost 87 99 (6 %)
AFUDC 278 306 (20 %)
1440 1592 (100 %)
6-62
-------
Table 3
Financial Assumptions
Source Of Capital
Capitalization
Ratio
Cost
Common Equity
45%
13.4%
Preferred Stock
10%
10.0%
Debt
45%
10.0%
Federal & State Income Taxes
38.0%
Book Life, Years
30
Tax Life, Years
20
Initial Year Of Design & Construction 1990
Beginning Of Commercial Operation 1995
Allocation Of PFI Over Design/Construction Yeras
15%, 25%, 35%, and 25%
PFI Escalation
General Inflation
Coal Cost/escalation
Methanol Credit/escalation
6.0% Per Year
6.0% Per Year
$45/ton, 6.8% per Year
$0.35/gallon, 9% per Year
6-63
-------
95% Purity
Oxygen
108 MW
Aux. Power
Net Power
~ To Grid
927 MW
67] MW
7.412 stpd
364 MW
Wuer
SannaJ-
ion
Gu
Turbines
are 1
Block Flow Diagram Showing an IGCC Plant Without Spare
Gasifier or Methanol Plant (Slurry-Feed Gasifier)
95% Purity
Oxygen
165 MW
Am. Power
704MW
Net Power
To Grid
993 MW
MeOH exnnder
~ 4 MW
Design
Coai
450 MW
Add
Gas
Removal
Onee-
ihroogh
Methanol
Plant
Acid
Gas
Removal
Gas
Turbines
and
HRSGs
Figure 2
Block Flow Diagram Showing an IGCC/OTM Plant
Slury-Feed Gasifier
6-64
-------
Percents
100,
IGCCw/o IGCCwiih IGCC with Combined
Spare gasifier Spare Gasifier Spare Gasifier Cycle Plant
or MeOl l Plant & MeOlI Plant with Natural
Gas
Figure 3 Plant Equivalent Availability
-------
r
IGCC base case
No MeOH production
37
MeOH @ $0.35'/gal
IGCC/OTM
base case
E
MeOH @ $0.40/gal
o 3 4
t5
_a>
LU
H—
o
(0
O
O
32
9
8
7
6
Methanol Price Escalation, %
Figure 4 Effects of Methanol price and Escalation on COE
-------
REFERENCE
1. Frank, M. E.f Methanol Production Scenarios. In Proceedings:
Fourteenth Annual EPRI Conference on Fuel Science, Palo Alto,
California. May 1990, pp. 18-1 to 14 (EPRI GS-6827)
2. Klosek, J. Sorensen, J. C., Energy Storage with Coal Gasification
Combined-cycle Electric Power. In Proceedings: Gulf Coast
Cogeneration Association Spring Regional Conference, Houston,
Texas. May 11,1990.
3. Mednick, R. L., Weatherington, R. W., Pech, J. Wv and Wright T. L.,
Design Criteria for Once-through Methanol Using the LPMeOH
Process. In Proceedings: Eleventh Annual EPRI Conference on Fuel
Science. Palo Alto, California. Mav 1986.
r j
4. Weber, W., Walters, A. B., and Tam, S. S. Hydrogen from Coal
Gasification Plants. Paper presented at the Third Annual U. S.
Hydrogen Meeting. Arlington, Virginia. March 18-20, 1992.
6-67
-------
Paper 6-F
EPA'S COST-SHARED SOLAR ENERGY PROGRAM
by: Ronald J. Spiegel
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
abstract
The objective of this program is to establish and demonstrate solar energy
costTShared commercialization projects to demonstrate how they can be used to
displace fossil fuels. The program will also have a major impetus to validate the
ability of solar energy to be used as a pollution mitigation technology. Further, the
demonstrations will assist in removing obstacles to the marketplace for solar
technologies by assisting in quantifying environmental concerns. This paper discusses
a project which has just commenced in the area of photovoltaic demand-side power
supplies. Additional discussion is provided relative to future projects which are being
contemplated.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency's peer and
administrative review policies and approved for presentation and publication.
6-66
-------
INTRODUCTION
Growing use of fossil fuels has generally contributed to environmental
degradation including such problems as acid rain and the greenhouse effect (the
buildup of global warming gases such as carbon dioxide, carbon monoxide, and
methane). It is, therefore, imperative to investigate technologies that have potential
to prevent or mitigate these environmental problems and also make sense from
energy, engineering, and economic viewpoints. Solar energy technologies meet these
criteria. Because of solar energy's huge potential and the clear environmental benefits
from its usage, EPA intends to establish and demonstrate the practical potential of
solar energy applications that are close to being commercialized and thereby provide
added impetus to their commercialization and, at the same time, utilize the
technology as a pollution reduction strategy or to comply with potential environmental
concerns.
The general goal of the program is to demonstrate the technical and economic
feasibility of solar technologies as a pollution mitigating energy replacement of fossil
fuels. The general approach is to install and monitor solar energy systems at
different geographic locations in the United States for users of retail electricity who
are interested in having such systems on their premises. These systems should supply
electrical power directly to the user, where they have to compete only with the retail
cost of electricity. These systems would be monitored for performance and reliability
using on-line remote monitoring all tied into a central facility at EPA's Air and Energy
Engineering Research Laboratory (AEERL) at Research Triangle Park, NC. Participants
in the demonstration projects will be required to have considerable experience in the
selected technology, to select an appropriate solar energy system, to determine the
requirements for the technical and economic aspects of the technology to be a
pollution mitigating energy strategy, to select appropriate host sites, to conduct the
required demonstration activities, and to cost-share at least 50% of the project (at
least 30% for companies qualifying as small businesses).
PHOTOVOLTAICS (AN EPA PROSPECTUS)
Within the solar area, perhaps the most promising and potentially ubiquitous
energy option is based upon photovoltaic (PV) conversion, the transformation of
solar radiation directly into electric power. Tremendous opportunities exists for solar
PV technologies to assist in meeting the energy needs of the 1990's and beyond. The
costs have come down dramatically (about 68% per decade) since solar cells were
first used in space. The current levelized cost of energy for PV is around 30 to
35c/kWh, with costs expected drop to the 4 to 7e/kWh range by the year 2010 [1].
Thus, in the domestic bulk power markets, head-to-head competition with gas- and
coal-fired plants before 2010 will be tough. However, even at today's costs there is
an untapped remote (off-grid power supplies) world market of perhaps 200-300 MW.
6-69
-------
For grid-connected applications, nearer term opportunities appear to be on the
customer (retail) side of the meter, where load growth could be met with PV demand-
side and distributed power supplies. For example, summer peaking rates can range
from 11 to 44c/kWh which would imply that for demand-side power supplies to shave
peaking loads on buildings could be cost effective today in some locations. The utility
companies or customers could own the on-site PV power generators.
Another near term possibility for grid-connected PV is distributed power
sources for standby generation and dispatchable power for the utility grid. For load
centers where demand growth has strained local transmission and distribution
capabilities, PV distributed power supplies could be an option to upgrading a
substation or constructing a new substation. In fact, it has been reported [2] that
current PV systems could be marginally competitive today in certain locations. For
these sites, avoided utility costs would accrue due to savings in equipment capacity
additions/upgrades (generation, transmission, and distribution) and in increased
reliability of customer service. That study also found that, while costs somewhat
exceeded benefits today, by 1995 the benefits provided by PV distributed systems
would exceed costs. This projection, of course, assumes that the levelized cost of
energy for PV will decline.
With the potential increase of electric vehicles (EVs) in the near future, solar PV
carports are expected to provide daytime recharging tor EVs. The primary
recharging stations for EVs will be the users' homes, with batteries charged at night.
These daytime recharging stations could also be grid-connected to provide solar-
generated electricity to buildings when the vehicles are not plugged-in. There is a
good match between the peak electricity provided by the recharging station and
daytime recharging needs of the EVs. Of course, the number of recharging stations
will depend on how many EVs are ultimately on the road iand how inexpensive PV units
become.
The future environmental impact of PVs is difficult to estimate at this time. The
U.S. Department of Energy (DOE) [1] has projected that PV market penetration in the
U.S. by the year 2010 could range from approximately 6 to 27 GW. Using a 27.5%
capacity factor for PV and offsetting emissions from electric power generation using
a coal-fired plant, it can be shown that for each gigawatt of PV generating capacity,
approximately 2.2 million tonnes less of carbon dioxide (CO2) is emitted to the
atmosphere per year. Thus, if these market projections are correct, by the year
2010 a yearly reduction of emissions of CO2 could be achieved by PV on the order of
13 to 59 million tonnes.
Thus, EPA's R&D efforts will tend to be in the PV systems application
development arena, where DOE traditionally has not allocated much funding. EPA's
future activities are likely to be in: demand side power supplies, distributed power
6-70
-------
supplies, integrated building applications, daytime recharging stations for EVs, and
advanced control systems for PV/hybrid systems.
CURRENT PROJECT (DEMAND-SIDE MANAGEMENT)
Demand-side management (DSM) activities by utilities have been around for a
while. Most utilities conduct their own DSM programs, but some have turned to
energy service companies to acquire load management capabilities. Usually, peak
demand reduction is their primary goal, but energy reduction has also been targeted.
However, most utilities have been somewhat reluctant to commit to demand side
power supplies, especially new and emerging technologies, as a means for load control
(DSM). To fill that void, EPA has embarked on a project to install and monitor PV
energy systems in several geographical locations in the U.S. for retail-end users of
electricity. The goal of the project is to investigate how PV technology may be used
as a pollution mitigating energy replacement of fossil fuels by reducing electrical
demand by commercial and residential buildings. The project requires that the PV
systems:
be located on the user's premises;
• supply electric power directly to the user;
reduce the electrical demand by commercial and/or residential buildings;
compete only with the retail cost of electricity;
be modular and replicable;
be capable of generating 5 to 15 kW at each site; and
obtain performance data over a complete heating and cooling season
and transmit the data to EPA.
In August, 1992, a contract is expected to be awarded to Ascension
Technology, Lincoln Center, MA, to conduct the study. Additionally, Ascension
Technology has 10 utility partners that will provide significant co-funding, along with in-
kind support. Seventeen PV systems (eight commercial buildings and nine residences)
will be installed and monitored at various sites in the U.S. The actual site selections
have not been finalized yet. The total PV power involved is 116 kW. Applying the
previously stated yearly emissions reduction potential of PV (1 GW reduces CO2
emissions by 2.2 million tonnes), this project will produce a reduction of 255 tonnes of
CO2 per year.
The total environmental benefits of DSM using PV power supplies could be
significant. For example, if 50 million residential users in the U.S. installed 4 kW rooftop
PV generators to produce daytime electricity, the electricity produced could displace
200 GW of central station generated electricity. This calculation assumes that this
offset electricity is produced from coal to estimate an upper value for the pollution
reduction potential of PV power supplies in a DSM role. Obviously, the offset
6-71
-------
electricity could be produced by a complicated mix of gas, oil, coal, and nuclear
plants. With this assumed, over 400 million tonnes less CO2 would be emitted yearly.
FUTURE PROJECT (PHOTOVOLTAIC - ELECTRIC VEHICLE RECHARGING STATIONS)
Electric vehicles (EVs) are rapidly becoming a focus of public attention [3].
Legislation has been passed in California and Massachusetts, and is pending in 10
other states, to require a specified percentage of automobile sales in these states to
be EVs by 1998. The impetus behind this legislation is that EVs offer the promise of
cleaner air to cities. In the United States today there are 185 million gasoline fueled
vehicles on the road with an additional 7 million added annually. Total vehicle
kilometers traveled doubled in the last two decades from approximately 1.6 trillion in
1970 to 3.2 trillion in 1990, and current estimates predict another doubling by the
year 2000. The number of cars on American roads and the kilometers they travel are
increasing much faster than the human population. Although the United States has
the most stringent emission regulations in the world, motor vehicles account for
approximately 35% of total volatile organic compound (VOC) emissions, 40% of total
nitrogen oxides (NO*) emissions, and over 65% of total carbon monoxide (CO)
emissions.
Adoption of EVs by the public would necessitate changing production (car
manufacturers) and buying (consumer) patterns. Several recent legislative policies
may provide catalysts for these changes. For example, the California plan does
mandate EVs: 2% of all cars sold in California in 1998, rising to 10% by 2003. If the
eight Northeast States for Coordinated Air Use Management adopt the California
plan, between 250,000 and 300,000 EVs could be on U.S. roads by 2003. Also,
several large automobile manufacturers (e.g., GM, Ford, Chrysler, Nissan, and BMW)
have major EV projects ongoing. The GM Impact has changed the vision of EV
performance to that of high performance sports cars [3]. It is anticipated that a
significant increase in EV sales could begin as early as 1995.
The environmental benefits from EVs are difficult to quantify at this time. A
study conducted by the South Coast Air Quality Management District [4] showed that
by replacing 5% of conventional gasoline fueled vehicles with EVs in the South Coast
Air Basin would reduce yearly VOC and NOx emissions by approximately 8950 and
6600 tonnes, respectively. However, California would probably benefit more than
other states because their electricity is produced primarily by natural-gas fired steam
plants. In other areas, there could be an increase in emissions associated with a
growth in electricity consumption and production. However, many of the power
plants may be located in less environmentally sensitive geographical locations, and if
EV recharging occurs at night, emissions will be produced at off-peak ozone forming
hours.
6-72
-------
It is generally assumed that most EV batteries will be charged at night when
demand for electric power is low. However, it is almost a certainty that some
infrastructure must exist for daytime recharging, especially for commuter vehicles. If
EVs are bought in large numbers, driving patterns are likely to be similar to internal
combustion vehicles, which will require daytime "refueling." EVs charged with powier
from PVs would truly be zero emission vehicles.
EPA is currently contemplating a PV recharging scenario for EVs. While no
decision has currently been reached, the likely approach will be to install two or three
PV recharging stations in non-attainment ozone metropolitan areas. One year's data
would be gathered to demonstrate that EVs and a PV recharging facility can be a
transportation option to reduce greenhouse gases.
CONCLUSION
EPA has embarked on a study to determine the environmental benefits potential
of DSM using PV. The R&D focus will be in the systems application development area
as a pollution reduction strategy.
REFBRBCES
1. The Potential of Renewable Energy: An Interlaboratory White Paper, U.S.
Department of Energy, SERI/TP-260-3674, DE 90000322, March 1990.
2. Benefits of Distributed Generation in PG & E's T&D System: A Case Study of
Photovoltaics Serving Kerman Substation, Pacific Gas and Electric Company, GM
663024-8, August 1991.
3. Fischetti, M., Here Comes the Electric Car - It's Sporty, Aggressive, and Clean,
Smithsonian, Vol. 23, pp. 34-43, April 1992.
4. Long Range Strategies for Improving Air Quality, South Coast Air Quality
Management District and the Southern California Association of Governments,
September 1985.
-------
PHOTOVOLTAIC DEVELOPMENTS
Jack L. Stone
Technical Director, Photovoltaic Programs
National Renewable Energy Laboratory
Golden, Colorado 80401
ABSTRACT
Photovoltaics, the direct conversion of sunlight to electricity, is an environmentally pristine,
renewable energy option currently used in a large number of international applications, principally non-
grid-connected in remote locations. Examples include water pumping, communications, vaccine
refrigeration, and village power. Over the past 15 years, a very aggressive research and development
program carried out by the U.S. Department of Energy has continued to improve photovoltaic performance
and reliability and has contributed to lower production costs. Along with these improvements have come
a large number of new, cost-effective uses, including larger power installations in a utility environment.
This technical and economic progress will be reviewed to represent the current status of this renewable
energy option. Near-term plans for large-scale installations in the United States will be discussed along
with the required costs for competitiveness with conventional electricity generation options.
i.
6-74
-------
Photovoltaics (PV) is the technology by which sunlight is converted directly into direct current
(dc) electricity. The technology utilizes many of the same techniques common to solid-state electronic
devices such as integrated circuits. PV involves no moving parts, so there is nothing to wear out. and it
is environmentally benign during operation. PV is modular in that individual solar cells are combined in
series and parallel combinations to synthesize the desired design voltages and currents (hence power) to
form multiple solar cells into what are known as modules. Multiple modules are further wired in series-
parallel combinations to form larger powered arrays. Using these various combinations, PV systems can
literally be built in sizes ranging from less than a watt to many megawatts. Depending on the application,
the dc power can be converted to alternating current (ac), allowing it to interface directly to the electric
utility grid.
INTRODUCTION
Figure 1 shows the basics of the solar cell. Essentially, the PV device is a battery like those
commonly used in many applications. It has two terminals but no consumables to be replaced. Instead,
the PV "battery" is fueled entirely by sunlight. The inherent energy of the sunlight is transferred to the
material from which the PV device is fabricated (typically a semiconductor such as silicon). The energy
is transferred through an absorption process. In the process, electrons and holes in the semiconductor are
liberated by the sunlight and are able to move under the influence of a built-in electric field created by
the positive/negative (pn) junction that is ftindamental to the operation of the device.
If all of the solar energy were
converted to electricity, each square
meter of PV cell would produce 1000
watts. The efficiency of conversion
is, however, limited by physical
processes to less than 100%.
Typically, PV devices convert from
about 5% to 35% of the available
solar energy depending on the type of
semiconductor material used. As will
be shown later, the higher the
efficiency of the module, the more
that can be afforded for the module.
There are two fundamental
types of PV systems. The first is the
"fiat plate" system, and the second is
the concentrator system. Hat-plate
systems are typically mounted in a
fixed position relative to the sun.
They can, however, be tracked to
follow the sun in both its hourly
movement and its seasonal change in position. The concentrator uses either mirrors or lenses to focus the
sunlight onto a small area. The concentration raises the power density by a factor of 10-1000 over a
typical flat-plate module. Both approaches have their own economical parameters that influence the
ultimate cost of PV-produced electricity. In addition, concentrators can be used effectively only ill
geographical areas where the so-called direct component of sunlight is plentiful. An example would be
the desert southwest in the United States. Typically, flat plates can be used in almost all geographic areas.
n-Type
semiconductor
Electrical
energy
p-Type
semiconductor
Photovoltaic device
Figure 1. In the typical PV cell, photon energy fives
electrical charge carriers, which become part of the current
in an electrical circuit. A built-in electrical field provides the
voltage needed to drive the current through an external load.
6-75
-------
THE SOLAR RESOURCE
There is a common misunderstanding that the United States has inadequate sunlight to allow
significant energy to be generated by photovoltaics. Actually, the energy available is about 10" kilowatt-
hours per year, about a million times as much energy as the United States consumes on a yearly basis.
The map shown in Figure 2 indicates that the average yearly solar resource varies from 1.6 megawatt-
hours per square meter per year at the least desirable regions to 3.8 megawatt-hours per square meter per
year in the best areas. Therefore, flat-plate collectors can operate well whether in Seattle or Phoenix. This
is because the solar spectrum consists of a direct and a diffuse component, and flat plates use die total
spectrum whereas concentrators can concentrate only the direct component. The direct component is
significantly reduced by clouds, dust, and other airborne particles. Therefore, concentraors will find
application primarily in the desert southwest, where the direct component is more available.
I
Megawatt-hours
per square meter
1.6-1.8
1.8-2.2
22-2.6
2.6-3.0
3.0-3.4
3.4-3.B
Figure 2. Average annual solar radiation for the United States
Sunlight has a color spectrum varying from the long-wavelength infrared to the short-wavelength
ultraviolet Depending on the semiconductor used to fabricate the solar cells, its sensitivity will vary with
the color of the spectrum. The maximum theoretical efficiency is limited to approximately 30%-35% for
flat-plate solar cells. Concentrated sunlight allows the cells to operate at higher efficiencies, perhaps as
high as 40%—45%.
ECONOMICS OF PHOTOVOLTAICS
For PV to be widely used, the costs must be competitive with those of conventional forms of
electricity. In the United States, average prices for electricity are 6-7 cents per kilowatt hour. Today's
PV generates electricity at 20-30 cents per kilowatt hour, so these costs must be reduced by about a factor
of five. A number of factors influence PV energy costs. Foremost are the module efficiency, lifetime,
and cost on an area basis. Several economic factors, principally the cost of money, also factor into the
6-76
-------
equation to determine the energy cost. The U.S. Department of Energy (DOE) chose, as a target, 6 cents
per kilowatt hour. Figure 3 indicates the module costs and module efficiency that combine to produce
the 6-cent goal. The figure assumes a module lifetime of 30 years. It is evident from the figure shown
for flat-plate systems, that lower efficiencies require lower module costs. In fact, module costs can be
traded off against efficiency. Similar curves exist for concentrators, which require higher efficiencies to
offset the higher costs of the required
trackers and lenses
concentrate the sun.
to follow and
150
100
in
O
o
£
3
¦D
O
2
50
I / / /
Module Efficiency /25% / /
/ y/20% /
Fixed Flat Plate j
/ /l5%
/ /io%
r- r ~i—
Planning Target
1 1
2 4 6 8 10 12
Leveiized Electricity Cost (#/kWh in constant S)
CURRENT MARKETS AND
APPLICATIONS
PV
Of
The worldwide market for
today is approximately 50 megawatts,
this amount, the United States has about a
one-third market share; the rest is sold
primarily by Japan and Europe. Much of
the market is concentrated in remote
applications and consumer products.
Remote applications include water pump-
ing, communications, vaccine refrigera-
tion. cathodic protection, and small village
home power. The technology used in
these applications is primarily crystalline
and polycrystalline silicon. Consumer
applications include small products such
as calculators, watches, and small battery
chargers. Most of these products axe
fabricated from amorphous silicon. In all
cases, the applications are considered to
be high value and are competitive with
the alternatives such as diesel power,
batteries, and kerosene. As the cost of
PV is reduced, more of these applications
will be available for PV to fill.
Eventually, larger applications will
become cost effective, including
residential and commercial building
demand-side-management projects and
The industry is currently caught in a "catch-22" situation. That is, as the
However, to justify the
And. lower costs are
Figure 3. Module costs and efficiencies vs. 30-year
leveiized electricity costs for fiat-plate photovoltaic
systems (fixed flat plate)
larger bulk power stations.
production plant sizes increase, prices are reduced through economy of scale,
capital expenditures for plant construction, sustainable markets must be identified,
necessary to generate these sustainable markets. Markets are currently growing at about 20% per year.
This evolutionary approach to market growth will delay the construction of the larger, cost-effective
production facilities. To accelerate this process. DOE instituted activities to identify cost-effective
applications within the electric utility industry that can be aggregated to form larger, sustainable markets.
Under consideration are applications such as PV in lieu of line extensions, system upgrades, and
transmission and distribution controls such as sectionalizing switches. As an example, power applications
that are more than one-half mile from the utility grid are many times less expensive when provided by
6-77
-------
PV/battery systems instead of by extending the utility grid. In fact, in Colorado, the utility regulators
require the utility to provide the customer with an analysis of the use of PV when the monthly kilowatt
hour usage divided by the distance to the utility line is less than 1000 (the so-called "Rule 31"). Most
planners forecast a massive growth in PV systems when the price of PV-generated electricity reaches about
10 cents per kilowatt hour.
STATUS OF PV TECHNOLOGIES
It is difficult to track the technical progress of PV. The most generally accepted figure of merit
is the device or module efficiency. Compounding the problem is the fact that different organizations
measure efficiency under a variety of test conditions. To allow a comparison of different technologies
between different laboratories, the National Renewable Energy Laboratory (NREL) provides efficiency
measurements under carefully controlled test conditions. The accompanying tables list the most current
results for a wide variety of materials and device configurations. The numbers represent the best devices
thai NREL has measured but may not represent the best reported in the technical literature. As indicated
in the tables, the measurement uncertainty is estimated to be +2% for single-junction cells and +5% for
multiple-junction cells.
Crystalline or polycrystalline silicon continues to be the material of choice for high-efficiency,
highly reliable PV modules. However, to reach the ultimate low costs, there is a growing consensus that
a thin-film technology will be required that uses less semiconductor material in highly automated
production facilities. Currently, industry is pursuing a variety of approaches and materials, including
amorphous silicon deposited by glow discharge, thin-film polycrystalline silicon grown from solution,
copper indium diselenide (CuInSeJ deposited by sputtering of the metal followed by in-situ selenization,
and cadmium telluride (CdTe) either spray-deposited or grown by close-spaced sublimation. All of these
techniques are making excellent progress and have a high probability of success. Several of the
approaches have been used to fabricate modules and are undergoing evaluation in a utility test
demonstration at the Photovoltaics for Utility-Scale Applications (PVUSA) site in Davis. California.
THE FUTURE
Future progress in the widespread use of PV can take several paths, depending on market
development or market pull scenarios. One possibility is a business-as-usual or evolutionary growth path
continuing to expand at 20% or more a year. Such a track will push off into the future, perhaps to the
year 2010, any serious investment in larger, cost-effective production facilities. The second track is a
more revolutionary market expansion through growth in competitive high-value applications, coupled
perhaps with government policy changes (possibly driven by environmental concerns), and U.S. utility
usage of PV. In the near term, the international market will continue to be a major user of PV. Consumer
interest in PV buildings applications (demand-side management) could also allow market expansion,
dearly, the technological progress to date indicates a technology readiness to meet market demand.
Economic factors, rather than technical ones, are now the major forces controlling PV's future.
6-78
-------
FURTHER READING
The interested reader is referred to the following list of publications for additional information.
The technical level varies from low to highly technical.
1. Fundamentals of Solar Cells, A.L. Fahrenbruch and R.H. Bube, Academic Press (1983), a very
complete reference book with a high degree of technical detail.
2. Solar Cell Device Physics, S.F. Fonash, Academic Press (1981), a highly technical coverage of
device physics.
3. Solar Energy Conversion: The Solar Cell, R.C. Neville, Elsevier (1978), a medium technical level
of the technology.
4. Solar Cell Array Design Handbook, H.S. Rauschenbach, Van Nostrand Reinhold (1980), a good
reference on systems-related issues.
5. Solar Cells. edited by CJE. Backus, IEEE Press (1976). a collection of historical technical paper
reprints.
6. Basic Photovoltaics Principles and Methods, Solar Energy Research Institute, Van Nostrand
Reinhold (1984), very general for readers without extensive technical background.
7. Harnessing Solar Power: The Photovoltaics Challenge, K. ZweibeL, Plenum Press (1990), a very
good introduction to PV technology.
In addition, three international conferences meet every 18 months (spaced 6 months apart) for
which proceedings are published. These range from highly technical to general overviews. The
conferences are the IEEE Photovoltaic Specialists Conference (the United States), the EC Photovoltaic
Solar Energy Conference (European), and the Photovoltaic Science and Engineering Conference (Far East).
ACKNOWLEDGEMENT
The author wishes to thank Mr. Keith Emery of NREL for furnishing his latest efficiency data for
inclusion in this paper. NREL is operated for the U.S. Department of Energy under Contract No. DE-
AC02-83CH10093.
The work described in this paper was not funded by the U.S. Environmental Protection Agency
and therefore the contents do not necessarily reflect the views of the Agency and no official endorsement
should be inferred.
6-79
-------
Amorphous Silicon
Source
Date
Total Area
(cm2)
Voc
(mV)
Jsc
(mA/cm2)
FF
(%)
Efficiency
(%)
Comments
ARCO
1/17/89
3.960
874
15.62
71.3
9.7
a-Si
Advanced PV Sys
1731/91
0.998
872
16.54
71.2
10.3
a-Si
Brookhaven
4/30/86
0.023
832
14.60
66.3
8.1
a-Si
Chronar
10/18/90
1.060
864
16.66
71.7
10.3
a-Si
1/4/88
0.980
2510
7.20
64.8
7.4
a-Si/a-Si/a-Si:Ge
3/20/89
1.000
1706
7.31
68.5
8.6
a-Si/a-Si:Ge
3/10/86
0.284
878
14.40
65.9
8.3
Photo CVD a-Si
ECD
2/16/88
0.268
940
15.20
69.4
9.9
ITO/a-Si/ss
6/13/89
0.269
1876
7.43
73.6
10.3
ITO/a-Si/a-Si/ss
2/16/88
0.268
1640
10.10
69.6
11.6
ITO/a-Si/a-Si:Ge/ss
2/16/88
0.268
739
19.80
59.6
8.7
ITO/a-Si.Ge/ss
¦
2/16/88
0.268
2641
6.96
70.0
12.4
ITO/a-Si/a-Si/a-Si :Ge/ss
IEC
11/7/87
0.284
862
17.60
66.8
10.0
Photo CVD a-Si
Glass Tech
5/4/90
1.017
862
17.26
69.8
10.1
a-Si (20A/s depos.rate)
9/15/89
0.986
886
17.46
70.4
10.9
a-Si
12/13/89
0.989
1711
8.45
71.6
10.4
a-Si/a-Si
3M
1/22/87
0.140
874
13.80
64.7
7.8
a-Si on Kapton
Solarex
4/16/87
1.077
879
18.80
70.1
11.6
a-Si
10/1/87
0.758
1685
9.03
68.1
10.3
a-Si/a-Si:Ge
6/16/89
0.256
2300
6.68
67.7
8.9
a-Si/a-Si/a-Si:Ge
Spire
12/9/86
0.099
878
16.60
72.2
10.5
a-Si
6/5/85
0.102
1686
6.90
57.4
6.3
a-Si/a-Si:Ge
All measurements performed at NREL, lOOOWm'2,26°C, ASTM E892 global unless noted otherwise. The area definition used for nonconcentrator
cells is the total area, the area for concentrators in the area designed to be illuminated. The total estimated uncertainty for Single junction cells is
±2%. Hie estimated uncertainty for multiple-junction cells is ±6%. This table was printed on Tuesday, July 14,1992. For fuHher information
contact Keith Emery at NREL [(303) 231-1032].
-------
Source
Date
Total Area
(cm2)
Voc
(mV)
Jbc
(mA/cm2)
FF
(%)
Efficiency
(%)
Comments
UPG
1/4/88
1.010
876
13.80
69.6
8.4
a-Si
Vactronics
11/11/86
0.256
822
11.40
60.5
5.7
a-Si
NREL
10/10/88
0.049
841
16.90
66.8
9.6
a-Si
All measurements performed at NREL, lOOOWm"2,25°C, ASTM E892 global unless noted otherwise. The area definition used for nonconcentrator
cells is the total area, the area for concentrators in the area designed to be illuminated. The total estimated uncertainty for single junction cells is
±2%. The estimated uncertainty for multiple-junction cells is ±5%. This table was printed on Tuesday, July 14,1992. For further information
contact Keith Emery at NREL [(303) 231-1032].
-------
AlGaAs, GalnP, GalnAsP, GalnAsGaAs, InP
Source
Date
Total Area
(cm2)
Voc
(mV)
^8C
(mA/cm2)
FF
(%)
Efficiency
(%)
Comments
ASEC
3/6/89
4.003
1035
27.57
85.3
24.3
GaAs, GaAlAs window
Kopin
11/13/89
4.000
1011
27.55
83.8
23.3
GaAs Cleft (separated)
6/3/88
1.000
1208
18.20
85.5
18.8
1.75 eV GaAlAs
3/12/90
3.910
1022
• 28.17
87.1
25.1
GaAs, GaAlAs window
4/16/90
16.00
4034
6.55
79.6
21.0
5mm Cleft GaAs, sub module
SMU
7/23/85
1.009
822
19.70
62.2
10.1
Thin Film GaAs/Ge/graphite
(direct)
Spire
3/6/89
0.250
1029
27.89
86.4
24.8
GaAs, GaAlAs window
4/21/92
16.14
1055
26.86
85.4
24.2
GaAs
3/7/89
0.250
1018
27.56
84.7
23.8
GaAs(MBE) Purdue
4/30/86
0.250
1200
16.20
82.6
16.0
1.75 eV GaAlAs
4/26/90
4.015
878
29.29
85.4
21.9
InP
4/24/87
0.249
1213
17.60
83.0
17.7
1.69 eV GaAsP
9/22/88
0.250
891
25.50
77.7.
17.6
GaAs on Si
11/25/88
0.250
1190
23.80
84.9
24.1
GaAs/Ge Tandem
Varian
3/8/89
0.500
2403
13.96
83.4
27.6
Two-terminal tandem
—
0.500
1402
13.92
86.8
16.0
1.93 eV GaAlAs top cell
—
0.531
1000
13.78
83.0
11.3
GaAs bottom cell
—
—
—
—
27.3
Three-terminal tandem
8/6/87
4.000
1244
16.00
84.6
16.8
1.75 eV GaAlAs
7/9/87
2.932
730
17.40
78.7
10.0
1.15 eV GalnAs bottom cell
under 1.75 eV GaAlAs
3/8/89
4.000
1045
27.60
84.5
24.4
GaAs, GaAlAs window
All measurements performed at NREL, lOOOWm 226°C, ASTM E892 global unless noted otherwise. The area definition used for nonconcentrator
cells is the total area, the area for concentrators in the area designed to be illuminated. The total estimated uncertainty for single junction cells is
±2%. The estimated uncertainty for multiple-junction cells is ±5%. This table was printed on Tuesday, July 14, 1992. For further information
contact Keith Emery at NREL [(303) 231-1032].
-------
Source
NREL
Date
Total Area
Voc
(mV)
«lsc
(mA/cm2)
FF
(%)
Efficiency
(%)
Comments
8/24/90
0.0634
973
1416
83.8
22.9
InP tope cell, 50 suns direct
—
0.0663
445
1321
75.7
JL2
0.75 eV GalnAs bottom cell, 60
suns direct
31.8
3-terminal InP/GalnAs tandem
10/11/90
0.0511
1096
990.3
83.6
23.1
GaAs top cell, 39.5 suns direct
—
0.0534
626
' 656.7
80.7
1A
0.95 eV GalnAsP under GaAs
top cell
—
—
—
—
—
30.2
4-terminal mechanically
stacked tandem
—
—
—
—
—.
—
39.5 suns, ASTM E891 direct
8/13/90
0.310
876
28.70
82.9
20.9
InP top cell
0.312
337
21.94
72.1
5.3
0.76eV GalnAs bottom cell
26.2
InP/GalnAs 3-terminal tandem
2/20/91
0.0746
899
6343
82.6
27.5
1.15 eV GalnAsP, 171 suns, direct
8/21/89
0.250
2292
13.61
87.4
27.3
GalnP/GaAs 2-terminal tandem
12/18/89
0.173
1038
28.70
86.4
26.7
GalnP/GaAs ,with aperture
9/13/89
0.249
1050
27.80
85.6
26.0
GaAs, GalnP window
1/18/91
0.0746
959
1509
87.3
24.3
InP homojunction 52 suns ,direct
12/10/90
0.0746
901
2588
79.3
21.0
Heteroepitaxial InP on GaAs,
88 suns, ASTM E891 direct
8/19/88
0.108
813
27.97
82.9
18.9
ITO/InP
4/11/86
0.100
1254
13.10
85.8
14.1
1.75 eV GaAsP
11/3/88
0.108
760
20.20
80.9
12.4
1.24 eV InAsP
10/14/90
0.160
658
531.8
82.0
9.4
0.95 eV GalnAsP under a GaAs
filter, 30.6 suns, ASTM E891 direct
All measurements performed at NREL, lOOOWm'2, 26°C, ASTM E892 global unless noted otherwise. The area definition used for nonconcentrator
cells is the total area, the area for concentrators in the area designed to be illuminated. The total estimated uncertainty for single junction cells is
±2%. The estimated uncertainty for multiple-junction cells is ±5%. This table was printed on Tuesday, July 14, 1992. For further information
contact Keith Emery at NREL [(303) 231-1032].
-------
Mono and Multi-Silicon
Source
Date
Total Area
(cm2)
V0C
(mV)
Jsc
(mA/cm2)
PP
(%)
Efficiency
(%)
Comments
EMC
9/3/85
4.100
692
31.50
79.1
14.8
Ribbon
Spectrolab
6/6/85
16.470
595
36.70
76.9
16.8
mono crystal Si
Spire
10/28/85
4.020
634
• 36.30
81.6
18.8
mono crystal Si
Stanford
6/3/86
9/20/88
0.162
8.625
681
702
41.00
40.70
78.4
77.7
21.9
22.2
Point-contact cell
mono crystal Si
Oak Ridge
10/3/86
4.000
667
36.00
81.6
19.3
mono crystal Si
Astro-Power
10/30/89
12/1/88
1.000
1.020
583
600
30.70
33.00
80.2
79.2
14.4
15.7
100 ^m thinned mono-Si
Thin film poly on ceramic
Sandia measurement
Westinghouse
6/8/85
11/29/84
4/11/86
1.017
24.600
1.004
602
598
637
31.60
30.40
33.70
80.7
79.5
76.2
16.4
14.5
16.5
Web; Direct, 28°C
Web; Direct, 28°C
mono crystal Si
Univ. of New
12/14/89
4.024
700
40.33
81.3
23.0
mono-crystal Si,aperture
South Wales
9/25/85
4/7/87
7/23/87
4.120
0.801
11.980
608
628
629
36.70
41.60
37.30
73.1
79.6
79.1
17.3
20.8
18.6
multi-crystal Si (Wacker)
mono-crystal: Cone,
mono-crystal: Large area
ISE Germany
10/16/91
4.023
675
39.67
77.8
20.8
mono-crystal
All measurements performed at NREL, lOOOWm"2,26°C, ASTM E892 global unless noted otherwise. The area definition used for nonconcentrator
cells is the total area, the area for concentrators in the area designed to be illuminated. The total estimated uncertainty for single junction cells is
±2%. The estimated uncertainty for multiple-junction cells is ±6%. This table was printed on Tuesday, July 14,1992. For further information
contact Keith Emery at NREL [(303) 231-1032).
-------
r
CuInSe2 and CdTe
Source
Date
Total Area
V0C
Jsc
FF
Efficiency
Comments
(cm2)
(mV)
(mA/cm2)
(%)
(%)
ARCO
6/17/88
2.400
464
38.50
65.4
11.7
ZnO/[Cd,Zn]S/CuInSe2
Boeing
12/30/86
1.074
509
30.40
66.0
10.2
CdS/CuInGaSe2
9/8/88
0.987
555
34.20
65.7
12.5
ZnO/[Cd,Zn]S/CuInGaSe2
1/2/92
3.990
532
• 31.88
67.9
11.5
ZnO/[Cd,Zn]S/CuInGaSe2
7/1/92
0.990
546
36.71
63.4
13.7
ZnO/[Cd,Zn]S/CuInGaSe2
IEC
9/2/87
1.028
445
35.00
64.6
10.1
ZnO/[Cd,Zn]S/CuInSe2
5/18/89
0.105
729
18.95
69.0
9.5
Glass/ITO/CdS/CdTe/Cu
10/22/91
0.191
790
20.10
69.4
11.0
Glaes/ITO/CdS/CdTe/Cu/Au
ISET
1/3/91
0.994
483
35.60
66.7
11.6
ZnO/CdS/CuInSe2/Mo/Glas8
NREL
6/12/86
1.033
446
35.30
65.3
10.3
CdS/CuInSe2
2/10/92
0.096
682
21.66
55.3
8.2
Glass/SnCVCdS/CdTe/Cu
Photon Energy
5/19/89
0.313
783
24.98
62.7
12.3
Glass/SnCVCdS/CdTe
5/2/91
0.300
788
26.18
61.4
12.7
Glass/SnCVCdS/CdTe
"SMU.Chu"
4/14/88
1.022
736
21.90
65.7
10.6
Glasa/SnCVCdS/CdTe/HgTe/Ag
U.S.Florida
5/24/91
1.197
840
21.93
72.6
13.4
GlasB/Sn02/CdS/CdTe/C/Ag
11/19/91
1.080
850
24.41
70.4
14.6
MgF2/Gla88/Sn02/CdS/CdTe/C/Ag
7/26/92
1.047
843
25.09
74.5
15.8
MgF2/GlaB8/Sn02/CdS/CdTe/C/Ag
AMETEK
10/19/89
1.068
767
20.93
69.6
11.2
Glaas/SnCVCdS/CdTe/ZnTe/Ni
All measurements performed at NREL, lOOOWm"2, 26°C, ASTM E892 global unless noted otherwise. The area definition used for nonconcentrator
cells is the total area, the area for concentrators in the area designed to be illuminated. The total estimated uncertainty for single junction cells is
±2%. The estimated uncertainty for multiple-junction cells is ±6%. This table was printed on Tuesday, July 14,1992. For further information
contact Keith Emery at NREL [(303) 231-1032].
-------
Source
Date
Total Area
V„c
Jsc
FF
Efficiency
Comments
(mV)
(mA/cm2)
(%)
(%)
Georgia Tech.
5/17/91
0.080
714
24.19
69.8
10.3
Glass/SnCVCdS/MOCVD
CdTe/ZnTe
6/28/91
0.080
745
22.10
66.0
10.9
Glass/SnOjj/CdS/MOCVD CdTe
U. Toledo
10/29/91
0.080
790
17.75
60.2
8.4
Glass/SnO^CdS/Laser ablated
CdTe/Cu/Au
All measurements performed at NREL, lOOOWrn"2, 26°C, ASTM E892 global unless noted otherwise. The area definition used for nonconcentrator
cells is the total area, the area for concentrators in the area designed to be illuminated. The total estimated uncertainty for single junction cells is
±2%. The estimated uncertainty for multiple-Junction cells is ±6%. This table was printed on Tuesday, July 14,1092. For further information
contact Keith Emery at NREL 1(303) 231-1032].
-------
Mechanically Stacked 4-Terminal PV Devices
Source
ARCO
NREL
Date Total Area
(cm2)
6/17/88 2.400
— 2.400
Boeing/Kopin 11/13/89 4.000
10/11/90 0.0511
— 0.0534
Voc
(mV)
Jbc
(mA/cm2)
FF
430
871
1096
626
17.10
15.62
1011 27.55
360 12.11
990.3
556.7
Efficiency
(%)
Comments
66.1
71.26
83.8
67.8
83.5
80.7
4.9
9.7
14.6
23.3
2.5
26.8
23.1
7.1
30.2
ZnO/[Cd,Zn]S/CuInSe2 under a-Si
a-Si top cell
4-terminal tandem: a-Si on
CuInSe2
GaAs Cleft stacked on
ZnO/[Cd,Zn]S/CuInSe2
4-terminal tandem, GaAs on
CuInSe2
GaAs top cell, 39.5 suns direct
0.96 eV GalnAsP under GaAs
top cell
4-terminal tandem GaAs on
GalnAsP
39.5 suns, ASTM E891 direct
All measurements performed at NREL, lOOOWm"2, 26®C, ASTM E892 global unless noted otherwise. The area definition used for nonconeentrator
cells is the total area, the area for concentrators in the area designed to be illuminated. The total estimated uncertainty for single junction cells is
±2%. The estimated uncertainty for multiple-junction cells is ±6%. This table was printed on Tuesday, July 14,1992. For further information
contact Keith Emery at NREL [(303) 231-1032].
-------
Summary of Thin-Film Amorphous Silicon Module Efficiencies
Cell Structure
Source
Date
Voc
lsc
FF
P max
Aperture
Efficiency
(V)
(mA)
(%)
(W)
Area (cm2)
. (%)
a-Si
ARCO
06/16/88
42.83
277
67
7.9
844
9.4
a-Si
ARCO
10/11/88
22.61
2102
71
33.6
4939
6.8
a-Si
ARCO
2/21/90
•
44.79
826
65
24.1
3962
6.1
a-Si
Chronar
3/10/88
23.87
1164
64
17.6
2626
6.7
a-Si
Chronar
10/23/89
67.98
1697
63
61.8
11344
6.2t
a-Si
Chronar
4/6/90
61.03
1848
66
74.4
11904
6.2tt
a-Si
Chronar
6/10/90
24.26
403
76
6.6
873
7.6
a-Si
APS
9/17/91
66.7
1980
70
77.8
12129
6.4ttt
a-Si
Glasstech Solar
10/16/89
22.14
368
66
6.4
880
6.1
a-Si
Glasstech Solar
10/16/89
12.01
2008
63
15.2
2703
5.6
a-Si
Solarex
4/6/87
26.38
410
68
7.1
1006
7.1
a-Si
Solarex
4/20/88
26.12
1362
66
22.5
3882
5.8
a-Si
Solarex
11/16/89
42.32
300
67
8.5
936
9.1
a-Si
Solarex
2/1/90
42.36
337
64
9.2
933
9.8
a-Si
Utility Power
10/16/89
16.08
479
64
4.9
750
6.5
Group
The module efficiencies are received prior to any light induced degradation
t Measured outdoors in Golden, CO, 1044 W/m2 total irradiance, 38°C module temperature
tt Measured outdoors in Golden, CO, 1004 W/m2 total irradiance, 33°C module temperature
tft Measured outdoors in Golden, CO, 1006 W/m2 total irradiance, 29°C module temperature
All measurements performed at NREL, 100 Wm-2, 26°C, ASTM E892 global unless noted otherwise. The area definition for modules is the
minimum rectangular aperture that can be placed over the module without afTecting the photo-current or the inside frame area. The total
estimated uncertainty for single junction modules is ±5%. The estimated uncertainty for multiple-junction modules is ±10%. This table was
printed on Tuesday, July 14,1991. For further information contact Keith Emery at NREL [(303) 231-10321.
-------
Summary of Thin-Film Multiple-Junction Module Efficiencies
Cell Structure Source Date Voc Ibc FF Pmax Aperture Efficiency
(V) (mA) (%) (W) Area (cm2) (%)
a-Si/a-Si
Advanced PV
Systems
1/28/91
48.8
183
66
6.8
942
6.6
a-Si/a-Si
Chronar
1/22/Q8
22.96
1119
69
16.1
2547
6.0
a-Si/a-Si
Chronar
1/30/89
47.76
191
65
5.9
842
7.0
a-Si/a-Si
Chronar
9/10/90
46.73
648
64
16.4
2555
6.4
a-Si/a-Si
Glasstech Solar
10/16/89
7.49
868
60
3.9
888
4.4
a-Si/a-Si
Solarex
9/06/88
48.78
193
61
6.8
1007
6.7
a-Si/a-Si
Sovonics
4/12/88
22.76
1668
67
23.9
3646
6.6
a-Si/a-Si
Sovonics
6/10/89
21.68
1671
67
24.0
3864
6.2
a-Si/a-Si
Sovonics
4/21788
22.20
2666
62
36.6
6634
5.5
a-Si/a-Si
Sovonics
4/12/88
22.76
1668
67
23.9
3646
6.6
a-Si/a-Si
United Solar
Corp. (USSC)
12/28/90
23.46
1883
61
27.1
3676
7.4
a-Si:C/a-Si:Ge
Solarex
10/24/88
49.86
234
60
7.0
1007
7.0
a-Si/a-Si/a-SiGe
Solarex
11/16/89
62.65
174
61
6.7
936
7.1
a-Si/a-Si:C/a-Si:Ge
Solarex
2/1/90
66.48
189
68
8.3
936
8. It
a-Si/a-Si/a-Si:Ge
Solarex
10/9/90
66.76
206
63
8.7
940
9.3
a-Si/a-Si/a-Si:Ge
Solarex
3/12/91
67.34
242
66
9.1
937
9.7
a-Si/a-Si/a-Si:Ge
Sovonics
10/6/88
17.01
666
64
6.0
838
7.2
a-Si/a-Si/a-Si:Ge
Sovonics
12/6/88
17.02
609
66
6.8
868
7.9
The module efficiencies are received prior to any light induced degradation
t Measured outdoors in Golden, CO, 1097 W/m2 total irradiance, 24°C module temperature
All measurements performed at NREL, 100 Wm-2, 25°G, ASTM E802 global unless noted otherwise. The area definition for modules is the
minimum rectangular aperture that can be placed over the module without affecting the photo-current or the inside frame area. The total
estimated uncertainty for single junction modules is 15%. The estimated uncertainty for multiple-junction modules is ±10%. This table was
printed on Tuesday, July 14,1991. For further information contact Keith Emery at NREL 1(303) 231-1032],
-------
Summary of Thin-Film CuInSe2 and CdTe Module Efficiencies
Cell Structure
Source
Date
Voc
f 8C
FF
P max
Aperture
Efficieni
(V)
(mA)
(%)
(W)
Area (cm2)
(%)
CdS/CdTe
AMETEK
10/24/88
9.63
126
67
0.68
100
6.8
CdS/CdTe
AMETEK
1/3/9P
7.05
118
49
0.40
103
3.9
CdS/CdTe
Photon Energy
3/23/88
21.07
395
67
4.8
808
6.9
CdS/CdTe
Photon Energy
10/24/88
20.49
519
57
6.1
838
7.3
CdS/CdTe
Photon Energy
11/2/89
20.76
463
58
5.4
930
5.8
CdS/CdTe
Photon Energy
9/3/91
~21
673
-55
6.7
832
8.If
CdS/CdTe
Photon Energy
3/26/92
35.6
893
50
15.9
3630
4.2tt
ZnO/CdZnS/CuInSe2
ARCO
5/30/91
24.01
244
64.4
36.7
3883
9.7
ZnO/CdZnS/CuInSe2
ARCO
1/23/89
23.04
2461
60
33.8
3985
8.5
ZnO/CdZnS/CuInSe2
ARCO
6/16/88
26.38
637
64
10.4
938
11.1
ZnO/CdZnS/CuInSe2
Boeing
12/3/86
1.78
774
63.9
0.88
97.0
9.1
f Measured outdoors in Golden, CO, 1013 W/m2 total irradiance, 33°C module temperature, under continuous
maximum power tracking to minimize voltage shift
-ft Measured outdoors in Golden, CO, 1063 W/m2 total irradiance, 22°C module temperature, under continuous
maximum power tracking to minimize voltage shift
All measurements performed at NREL, 100 Wm-2, 26°C, ASTM E892 global unless noted otherwise. The area definition for modules is the
minimum rectangular aperture that can be placed over the module without affecting the photo-current or the inside frame area. The total
estimated uncertainty for single junction modules is ±6%. The estimated uncertainty for multiple-junction modules is ±10%. This table waB
printed on Tuesday, July 14, 1991. For further information contact Keith Emery at NREL ((303) 231-1032].
-------
Concentrator Cell Efficiencies Verified at Sandia with Respect to the Standard
Concentrator Reporting Conditions
Source Illuminated Area Efficiency Concentration Materials System
(cm2) (%) (suns) .
BssaMasnaBBHaBnBBaMMHaBnaBHHBMHBnHBBHBMMnaBnaBnneBBSsaBaBBs
Univ. New South Wales
20.0
22.6
20
float-zone Si
Solarex
38.4
20.2
20
Czochralski Si
Solarex
39.6
17.6
20
multi-crystalline Si
Astropower
39.6
17.8
20
Czochralski Si
Stanford Univ.
0.15
28.2
140
float-zone Si
Univ. New South Wales
1.68
26.2
126
float-zone Si
SERA
0.066
23.9
65
float-zone Si
Solarex
1.68
21.6
150
float-zone Si
Varian
0.126
28.1
403
GaAs
Varian
0.126
29.2
206
GaAs
Spire
0.317
28.7
200
GaAs
Sandia, Varian and Stanford
0.317
31.0
600
GaAs stacked on Si,
4-terminal tandem
Boeing
0.063
34.2
100
GaAs stacked on GaSb
4-terminal tandem
All measurements performed at NREL, 100 Wm-2, 25°C, ASTM E892 global unless noted otherwise. The area definition for modules is the
minimum rectangular aperture that can be placed over the module without affecting the photo-current or the inside frame area. The total
estimated uncertainty for single junction modules is 16%. The estimated uncertainty for multiple-junction modules is ±10%. This table was
printed on Tuesday, July 14,1991. For filrther information contact Keith Emery at NREL [(303) 231-10321.
-------
Paper 6-H
ADVANCED ENERGY SYSTEMS FUELED FROM BIQMASS
by: Carol Ft. Purvis and Keith J. Fritsky
U.S. Environmental Protection Agency
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
ABSTRACT
The concentration of carbon dioxide (C02), a greenhouse gas. is increasing by
an estimated 0.5 percent per year. COz emissions from fossil fuel combustion
quadrupled between 1950 and 1980. Conversion of renewable biomass to energy is
C02 neutral and produces lower sulfur dioxide (SOz) and nitrogen oxides (NO,)
emissions than fossil fuel combustion.
The U.S. Environmental Protection Agency/Air and Energy Engineering
Research Laboratory is studying two biomass conversion technologies: conventional
combustion in a boiler coupled with a steam turbine system and gasification in a
gasifier coupled with an aeroderivative turbine system. In-house research is
addressing the problems encountered in conventional systems with regard to
emissions, tuoe fouling, bed agglomeration, and low thermal efficiency. Extramural
research is addressing the problems of advanced systems with regard to
fixed/fluidized-bed gasifiers, alkali/particulate cleanup, gas compatibility with turbines,
and system efficiency. The results will provide data for owner/operators to improve
system performance and for designer/developers to demonstrate advanced systems.
This research will help promote biomass-for-energy as a global warming mitigation
strategy by focusing on a need to maximize biomass resources through increased
utilization efficiency.
This paper has been reviewed in accordance with the U.S. Environmental Protection Agency s peer and
administrative review policies and approved for presentation and publication.
6-92
-------
INTRODUCTION
The leading culprit in the greenhouse gas arena is carbon dioxide (C02), and its
concentration is increasing by an estimated 0.5 percent per year [1]. Anthropogenic
activities are responsible for the unwanted buildup of C02 in the atmosphere. C02
emissions from fossil fuel combustion quadrupled between 1950 and 1980 alone;
therefore, reducing fossil fuel COz emissions is an obvious place to start mitigation
activities [1]. The efficient use of biomass is C02 neutral and can reduce pollution by
lowering sulfur dioxide (SO?) and nitrogen oxides (NOJ emissions. Figure 1 is a
simple representation of the biomass closed carbon cycle and the fossil fuel (oil, coal,
and gas) open carbon cycle.
Biomass is the principal single source of energy for 75 percent of the world's
population [2], It provides about 14 percent of the world's energy; 35 percent of the
total energy supply in developing countries, and 3 percent of the total energy supply in
developed countries (see Figure 2) [3]. This energy contribution is equivalent to 35
million barrels (5.6 x 109 L) of oil per day globally [2]. Some analyses suggest that the
amount of energy contributed by biomass could be increased by a factor of 13 [2]. By
the second quarter of the 21st century, biomass could provide 25-35 percent of the
total global power generation capacity [4], In 1990, the United States had 9,000 MW
of biomass-fueled electric capacity, 36 times the capacity in 1980, due to the Public
Utilities Regulatory Policies Act of 1978 (PURPA) [4J.
CLOSED
CAnBON
CYCLE
OPEN
CARBON
CYCLE
Figure 1 - Carbon Cycles
6-93
-------
FOSSIL FUELS 75%
A
FOSSIL FUELS 86%
HYDRO 6%
NUCLEAR 5%
BIOMASS 14%
WORLD
FOSSIL FUELS 58%
HYDRO 6%
NUCLEAR 5%
BIOMASS 3%
HYDRO 6%
NUCLEAR 1%
DEVELOPED COUNTRIES
BIOMASS 35%
DEVELOPING COUNTRIES
Figure 2 - Energy Use Distribution
Increasing biomass availability on a sustainable basis may not be viewed as a
high priority in developing countries and what is appropriate in one country might not
work as well in another. But energy demands will increase in all countries and all
countries have forestry and agricultural wastes and the potential tor energy crops.
Therefore, greater use oi industrial wood wastes, agro-processing wastes, and
agricultural residues will have tne greatest near term impact. The potential for the use
of c^sting biomass resources in the world is 65,381 x 106 GJ; 41,068 x 10€ GJ in
cteveUoing countries and 24,313 x 10s GJ in developed countries (3J. The
development and commercialization of small, residue-lueled facilities should fill the
neeas of rural communities and biomass industries as well as developing countries.
Promotion of carbon sequestration and biomass utilization requires research on
various issues including land use, forest management, production, harvesting,
reforestation, urban reforestation (carbon sequestration and mitigation of the urban
heat island effect), displacement of fossil fuels with biomass-based t sctricity and
liquid fuels, and increased use of wood products. The U.S. Department of Agriculture
(USDA)ZSoil Conservation Service, USDA/Forest Service, U.S. Department of Energy
(DOE)/Office of Conservation and Renewable Energy, U.S. Environmental Protection
Agency/Air and Energy Engineering Research Laboratory (EPA/AEERL), National
Renewable Energy Laboratory, and Oak Ridge National Laboratory are some of the
domestic organizations studying these issues.
6-94
-------
CONVERSION TECHNOLOGIES
In ethanol produciion from woody energy crops, 75 percent of the energy
ultimately produced can actually count toward fossil fuel displacement. When woody
crops are directly converted to electricity, this figure can increase to 90 percent [1].
This 15 percent increase indicates that the conversion technologies with the greatest
near term potential will focus on direct biomass-to-energy production. This near term
production should occur through utilization of residues in small stana-alone energy
production facilities.
EPA/AEERL's goal is to promote activities to achieve renewable production,
provide modern energy carriers, develop high-efficiency conversion processes,
develop high efficiencies for end use, and develop technologies that have favorable
economics on a small scale.
EPA/AEERL's program is studying two conversion technologies: conventional
combustion in a boiler coupled with a steam turoine system and gasification in a
gasifier coupled with an aeroderivative turbine system (see Figure 3).
EXHAUST EXHAUST
Combustion
Throughout the world, process heat/steam and electricity generated from
biomass fuels are almost entirely a result of direct combustion. Direct combustion
technologies fueled with biomass include dutch ovens, spreader-stokers, total or
suspension fired combustors, and fluidized or circulating bed combustors. Since these
technologies are firmly established, they are readily transferable to biomass energy
applications throughout the developed and developing world.
In applying these technologies to biomass fuels, problems are often
encountered with regard to unacceptable, uncontrolled emissions, boiler tube fouling,
bed agglomeration, and less than optimum thermal efficiency. These problems arise
as a result of the combustor's inability to respond to rapid variations in certain
properties of the fuel feed (moisture content, particle size, and inorganic content).
GENERATOR
CONVENTIONAL
ADVANCED
Figure 3 - Cycle Flow Diagrams
6-95
-------
In-house research at EPA/AEERL is addressing the above problem, and by
doing so. will provide data to owner/operators of biomass combustion systems for the
purpose of improving system performance through modifications to the combustion
process. This research will help promote biomass-for-energy as a global warming
mitigation strategy by focusing on a need to maximize biomass resources through
increased utilization efficiency.
The approach of the research is to burn different biomass fuels (sawdust
briquettes, bagasse pellets, and switch grass briquettes) in a pilot-scale stationary
grate combustor. A first series of tests is to use fuels with relatively uniform
properties. Standard thermal and chemical analyses will be performed to quantify fuel
properties. Data obtained from this series of tests will establish a baseline and
characterize relationships betwpen fuel properties and system performance. In a
second series of tests, fuel properties will be measurably altered (by increasing
moisture content or inorganic fraction, for instance) and the system will be monitored
and adjusted to op: mize combustor performance under adverse operating conditions.
The purpose of performing ir.is series of tests will be to minimize emissions of criteria
and non-criteria pollutants -- carbon monoxide (CO), NOt, particulate matter (PM), and
volatile organic compounds (VOCs) -- and fouling of heat transfer surfaces while
maximizing thermal efficiency by controlling air flows, combustion air temperature, etc.
Techniques for controlling the combustion process will be developed so that they can
be applied by operators of biomass combustion systems. Testing is anticipated to
begin in June 1993 when construction of the combustor facility is complete.
Gasification
Gas turbines fueled with biomass gas offer a great advantage in the 10 'o 50
MW range. Net plant conversion efficiencies could exceed 50 percent, especially if
the turbine exhaust is used in a heat recovery steam generator or an air bottoming
cycle (4). Aeroderivative gas turbines (i.e., gas turbines derived from aircraft jet
engines) offer high efficiency, low unit capital costs at modest scales, and low
maintenance due to their compact modular nature.
Fixed-bed Gasifiers
Temperatures of the gas exiting the gasifier are expected to be 500-600°C for
the fixed-bed updraft design. Within this temperature range, the alkali compounds
(formed primarily from potassium and sodium in the feedstock) appear to condense on
PM and can be controlled by particulate collection devices. The lemperature is high
enough that the tars formed will remain in the vapor phase and actually boost the
heating value of the gas. It would be desirable to have the gasifier and gas turbine
clcje-coupled to eliminate any condensation problems [4J.
The State of Vermont was funded by EPA/AEERL, DOE, and U.S. Agency for
International Development to evaluate the compatibility of gasified biomass feedstocks
with an aeroderivative gas turbine power generation system. The EPA/AEERL is also
6-96
-------
funding Vermont to perform a feasibility study on demonstrating this technology. The
compatibility evaluation was performed at the General Electric Corporate Research
and Development (GE CR&D) coal gasification pilot plant. The objectives of the study
were:
- to determine the composition of the product gas and gasification rate,
- to determine the nature of the contaminants in the gas that might have a
detrimental effect on the gas turbine,
- to measure the contaminants in the gas that might have a negative
environmental effect, and
- to determine the effectiveness of cyclones for particulate removal.
The gasification plant consisted of a feed system, a fixed-bed updraft gasifier,
and a cyclone (see Figure 4). The State of Vermont provided 83.8 tons (76 tonnes) of
dried wood chips and Winrock International provided 42.5 tons (39 tonnes) of bagasse
pellets.' These highly reactive biomass fuels were gasified at 20 atm (2.03 MPa).
The biogas product had a higher heating value than coal gas and was compatible with
gas turbine combustors. The particulate carry over and the alkali metal contained in
the particles indicated that a single cyclone was not sufficient for cleanup. The sulfur
emissions were lower than for coal combustion facilities equipped with flue gas
desulfurization systems, and fuel bound nitrogen levels were lower than for coal [5].
Fuel
Fuel
Primary
Cyclone
Qm
Dual
Figure 4 - GE CR&D's Equipment Flow Diagram
'Pelletized switch grass was the third luel considered bul the pellets crumbled in shipping. Briquettlng
Marketing and Services, Inc. was able lo produce a switch grass briquette per the GE size specification
on a demonstration briquettor. This demonstration size briquetlor had a small throughput that would not
produce the quantity required tor the test in the time allowed prior to testing.
6-97
-------
Fiuidized-bed Gasifiers
The most promising long-term prospect for biomass gasification is the use of
fluidized-bed gasifiers that can accommodate a wide variety of feedstocks. Fiuidized-
bed gasifiers have higher throughput capabilities and greater fuel flexibility than fixed-
beds, including the ability to handle low-density feedstocks like undensified crop
residues or sawdust. Gas quality control with fluidized-bed gasifiers may require
different treatment for two reasons. First, the exit gas temperatures are higher, 800-
900°C, which may result in the vaporization of alkalis. Second, there is more
paniculate carry-over, which cyclones alone will not handle, that will require ceramic or
sintered-metal barrier filters [4],
EPA/AEERL plans to cooperate with a researcher in the development of a
system consisting of a pressurized feed system, a fluidized-bed gasifier, and a filter for
gas cleanup. All components of the system have been designed to operate on
biomass fuel with minimal preparation and produce a gas suitable to fuel an
aerodenvative turbine. Testing of the system will begin in late '92.
REFERENCES
1. Trexler, M.C. Mindino the Carbon Store: Weighing U.S. Forestry Strategies to
Slow Global Warming. World Resources Institute, Washington, DC, 1991.
2. Hall, D.O., and Woods, J. Biomass: Past, Present and Future. In: Proceedings
of 1he Conference on Technologies for a Greenhouse-Constrained Society. Lewis
Publishers, Chelsea, Michigan, 1992.
3. U.S. Congress, Office of Technology Assessment, Fueling Development: Energy
Technologies for Developing Countries. OTA-E-516 (Washington, DC: U.S.
Government Printing Office, April 1992).
4. Williams, R.H., and Larson, E.D. Advanced Biomass Power Generation - The
Biomass-Integrated Gasifier/Gas Turbine and Beyond. |n: Proceedings of the
Conference on Technologies for a Greenhouse-Constrained Society. Lewis
Publishers, Chelsea, Michigan, 1992.
5. Furman, A.H., Kimura, S.G., Ayala, R.E., and Joyce, J.F. Biomass Gasification
Pilot Plant Study. Draft report prepared for the State of Vermont by GE Corporate
Research and Development. Schenectady, New York, 1992.
6-98
-------
Stephen Morgan
Citizens Conservation Corporation
Boston, MA
PROGRAMS AND POLICY IMPACTS ATTRIBUTABLE TO REGIONAL
BIOMASS PROGRAM WOOD STOVE RESEARCH EFFORTS
ABSTRACT
The decision by the CONEG Policy Research Center in the spring of 1985 to embark upon a
field evaluation of wood stoves led to an ambitious and sophisticated six-year research
program which has had profound impacts on the evolution of the industry. The initial
interest in exploring the claims of manufacturers about the efficiency, particulate emissions,
and creosote build-up in catalytic, catalytic add-on, and noncatalytic "high tech" woodstove
models readily expanded to undergird much of the conceptual thinking which informed the
EPA's participatory regulations negotiations process engaging the industry, the agency and
environmentalists. Since the publication of the EPA certification standards in 1987 and the
laboratory testing methodology utilized to support the standard, the NRBP-sponsored
research has moved on to test the field durability of certified stoves. The early,
disappointing findings in that field testing identified structural, design and materials flaws
which have directly contributed to improved stove designs in the second generation of
certified stoves.
The story of the Regional Biomass Program research is punctuated by these impacts:
1. The equipment and field testing methodology developed for particulate emissions
in the original study set the standard for all field testing in North America for the next
half-decade.
2. The findings in the original studies in the Northeast and Northwest influenced at
least three specific aspects of the New Source Performance Standard (NSPS)
established for wood stoves in 1987: the separate emissions levels for catalytic and
noncatalytic stoves; the burn rate and fuel cycle conditions for testing; and the
enforcement procedures related to physical inspections of the newly manufactured
stoves.
3. Unprecedented collaboration among US and Canadians in embracing this research,
improving the quality and aiding comparability of testing in both countries.
4. Successful leveraging of funding from a second federal agency, two state agencies,
and the manufacturer's trade associationfor four-six successive research projects.
5. The assembling of Project Advisory Committees of researchers, government
officials, regulators, and manufacturers for a frank exchange on both research and
policy issues.
6. The direct impact on stove designs following a public meeting with stove
manufacturers during the summer of 1989.
7. The new emphasis in the industry and concern among regulators about durability of
stoves certified since 1988; and the development of a reliable research instrument and
testing methodology to simulate field results are direct impacts of the research.
The work described in this paper was not funded by the U.S. Environmental Protection Agency. The
contents do not necessarily reflect the views of the Agency and no official endorsement should be inferred.
6-99
-------
Introduction
Over the last twelve years, significant amounts of research and dialogue have been
undertaken regarding wood stove technology and its environmental impact. Beginning with
the first study in 1982 in Ohio and New York, presented by Stockton G. Bamett of OMNI
Environmental Services to the Energy from Biomass and Wastes Symposium, concerns were
raised regarding wood stove performance and related particulate emissions, fluepipe creosote
accumulation, and overall efficiency. These concerns were particularly evident in areas of
the United States, such as the Pacific Northwest, where wood stove emissions contributed a
large percentage of environmental particulate emissions, heightening additional health and
safety concerns simultaneously as the wood stove manufacturing industry experienced
tremendous growth. Ski areas in Colorado and river towns in New England subjected to air
inversions also experienced problems with woodbuming.
In early 1984, the state of Oregon promulgated the first wood stove particulate
emissions regulations in the nation. These regulations allowed a phase-in period and set
separate standards for catalytic and non-catalytic stoves based primarily on laboratory testing
results. Two benchmark field studies were launched between 1985-1987: 1) the 1985-87
Northeast Cooperative Wood Stove Study (NCWS), often referred to as the "CONEG" study
(Coalition of Northeast Governors)1; and 2) the 1986-87 Whitehorse, Canada study. While
the final results of both of these studies were not available to the U.S. Environmental
Protection Agency until late 1988, the EPA was actively involved in setting up a
"regulatory/negotiation" (commonly referred to as "reg/neg") process to seek input and
comments from a wide range of industry, research, and government actors on proposed
regulations to limit particulate emissions within the wood stove industry by setting
certification standards for new equipment The interim results from the NCWS and
Whitehorse studies contributed significantly to the conclusions and discussions in the
"reg/neg" process. Since the promulgation of these regulations, phased in over two periods
between 1988 and 1990, additional continued testing of wood stoves has been conducted to
compare laboratory and field test results and to identify equipment degradation problems
and operator practice issues.
Several factors indicate the importance of both the research and its management on a
number of key areas within the evolution of wood stove manufacturing:
1) the initial research goal of improving wood stove design has led to a significant
amount of available data on different emissions impacts of variously designed
stoves, and the technology transfer resulting from this research has led to design
improvements in the wood stove manufacturing industry;
2) the testing methodology developed for this research, the "Automated Wood
Stove Emission Sampler" (AWES) and a data-logger, has set the industry
standard for such testing throughout the past ten years, and has most recently
been adapted for masonry heater and fireplace emissions studies;
'The Northeast Cooperative Wood Stove Study is commonly known as the "NCWS" study,
the "Northeast Study", and the "CONEG" study. Since much of the research literature and public
awareness of this study refers to the "CONEG" study, this terminology is widely used in this
report.
6-100
-------
Regional Biomass Program Impact
Page 2
3) the process used in planning and monitoring the early wood stove studies
brought together a unique group of industry and government representatives
prior to the promulgation of regulations, leading to important regulatory
compromises that produced a sense of livable" regulatory action;
4) the testing methodology and results led to active interest and cooperation
between two national governments (the U.S. and Canada) and among a number
of different US. government agencies and industry representatives; and,
5) continued testing of laboratory/field conditions raises future concerns not only
about stove design but also about the importance of operator-controlled
conditions and the on-going need for consumer education and training by
manufacturers and industry representatives. The current focus of testing on
stove durability issues has stimulated a creative tension between manufacturers,
who fear additional regulations, and other parties concerned about a steady
increase in emissions as certified stoves age.
Summaries of Significant Research Efforts
• The CONEG Study, 1985-87
The first major study of wood stove emissions and efficiency was conducted for the
Coalition of Northeast Governors (CONEG) Policy Research Center, and sponsored by
CONEG, the New York State Energy Research and Development Authority (NYSERDA), and
the U.S. Environmental Protection Agency (EPA). This study, known as the CONEG study
or the Northeast Cooperative Wood Stove Study (NCWS), was begun in the spring of 1985
and continued in various phases throughout the next six years.
The purpose of the first CONEG study, conducted by OMNI Environmental Services
of Beaverton, Oregon (the firm originally involved in the laboratory testing for the state of
Oregon DEQ), was to determine the effectiveness of both catalytic and non-catalytic new .
technologies in reducing wood use, creosote build-up, and particulate emissions. Sixty-eight
homes in Waterbuiy, Vermont, and Glens Falls, New York, were involved in the testing over
two heating seasons, 1985-86 and 1986-87. The study measured emissions, creosote build-up,
efficiency, and wood use for both conventional technology wood stoves, which provided the
baseline, and the new technology stoves. The study was the first major field research
activity to determine if field performance of the new technology could out perform
conventional technology and meet laboratory expectations.
An important facet of the CONEG study was the field introduction of a sampling
technology that would set the standard for almost all future wood stove emissions testing.
OMNI Environmental Services developed the Automated Wood Stove Emissions Sampler
(AWES) specifically to sample residential wood stove emissions. This sampler draws flue
gases through a probe which samples from the center of the flue pipe one foot above the flue
collar. The sample then passes through an XAD-2 resin trap that collects semivolatiles and
remaining particulates. Flue gas oxygen concentrations, used to determine flue gas volume,
6-101
-------
Regional Biomass Program Impact
Page 3
are measured by an electrochemical cell.2 The sampling period can be varied. The
promised development of the AWES by OMNI was a major reason the firm was selected
over five competitors when the CONEG contract was awarded in the late spring of 1985.
The AWES was accompanied by the "Data LOG'r", a programmable
microprocessor/controller capable of measuring a large number of variables including date
and time for recording periods, daily date and time in five-minute intervals, flue gas
averages over five-minute periods, ambient temperature in fifteen-minute intervals, wood
weights, oxygen measurements when the AWES unit was sampling, and AC power status,
measured at five-minute intervals (see Exhibit 1 and Exhibit 2 for a pictorial representation
of the AWES and data LOG'r systems).
After obtaining the baseline conventional stove data, three advanced technology stove
types were tested: catalytic stoves, add-on or retrofit devices (attached to conventional
airtight stoves at fluepipe), and low-emission non-catalytic stoves. Special emphasis was
placed on testing the effectiveness of the catalytic combustors.
Results of the CONEG study were published in November, 1987. Generally, the
CONEG study did indicate that the low-emission non-catalytic stoves and the catalytic stoves
had lower emissions and creosote accumulation than the conventional stoves. However, the
range of performances and the number of variables involved prevented definitive
conclusions about stove design and performance. In all cases, however, the advanced
technology episodically demonstrated lower emissions under the field conditions. The
average emissions rates for the stove types were 20.1 g/h for conventional stoves, 17.9 g/h
for retrofits, 16.4 g/h for catalytics, and 13.4 g/h for the low-emission non-catalytic stoves.
Unfortunately, while the new technology stoves did show decreased emissions, the
reductions were not as great as anticipated given the laboratory testing. The study was
unable to point to any single factor dominant in all stoves that affects emissions.
The CONEG study did raise several important issues: 1} it indicated the potential for
overall emissions reductions with new technology stoves; 2) it identified the issue of catalyst
performance/degradation over time (a major issue in the next round of testing); 3) it set the
standard accepted methodology and technology for field testing; 4) it raised the question of
operator practice as a potential factor affecting stove performance; and 5) it pointed out the
need for continued field and laboratory testing of both design and performance.3
2 This description of the OMNI AWES and Data LOG'r systems is compiled from the
numerous OMNI wood stove reports cited in the reference section of this report.
3 A year after the CONEG study began, the Northwest Regional Biomass Program
conducted its own complementary research in Portland, Oregon. Since the findings from
these studies were presented after the EPA regulations were published, they are discussed in
Part B.
6-102
-------
Regional Biomass Program Impact
Page 4
• The Whitehorse, Canada Study, 1986-87
In 1986-87 the City of Whitehorse, Yukon, Canada, conducted a study funded by the
Canadian Department of Energy, Mines and Resources. The findings of the study were
presented in September, 1987. The study was designed to assess the effectiveness of new
stove technologies in reducing emissions, reducing the risk of chimney fires, and improving
overall efficiency. The new technologies studied included integral catalytic wood stoves,
catalytic "add-on" retrofit devices, and low-emission non-catalytic stoves. In addition to
undertaking the testing as a means of determining whether laboratory predictions could be
fulfilled in the field, the study evaluated operator practices in the home, type of wood, and
flue pipe creosote build-up.
This study used the same AWES, Data LOG'r equipment as was used in the CONEG
study. Emission samplings were taken over a nine-week period in fourteen homes; four one
week sampling periods involved conventional stoves, and five one-week sampling periods
involved the new technology stoves.
While the basic types of stoves tested in the Whitehorse study were similar to those
tested in the CONEG study, the actual models differed. However, the results of the
Whitehorse study were similar to those of CONEG: conventional wood stoves exhibited
particulate emissions of 21.8 g/h, catalytic retrofit stoves averaged 16.2 g/h and integral
catalytic stoves averaged 12.1 g/h, and low-emission non-catalytic stoves averaged 143 g/h.
These are emissions levels more than twice as high as those achieved in laboratory
certification tests. While the CONEG study could not be statistically verified among the
technologies with a significant confidence level, the Whitehorse study did prove at a
statistically significant level that all of the advanced technology categories performed better
than the conventional stoves (even though the differences among the various advanced
technology categories were not as apparent).
The Whitehorse Study also pointed to major differences in creosote build-up in sealed
and vented double wall flue pipes. Within three days of the testing, the vented type of
double wall flue pipe was clogged with creosote and emissions rates actually increased.
With the sealed pipe, emissions decreased and the creosote problem was abated.
Additionally, the Whitehorse Study indicated a possible link between chimney length and
high emissions (straight, short chimneys with a mix of flue diameters may cause higher
emissions), and raised the importance of firebox size and bum rates in emissions testing.
Two key factors raised in the CONEG study were corroborated by the Whitehorse
study: 1) the hypothesis of the CONEG study that the advanced technology stoves could
perform better than conventional stoves in the field, providing better emissions control and
efficiency, was confirmed statistically in the Whitehorse study; and 2) both studies indicated
that field results of the advanced technology stoves did not match the Oregon certification
levels, even when using stove models certified in laboratory testing to those levels.
Both studies also raised issues of equipment degradation and operator influence as
being important factors in emissions control. As neither of these issues could be verified
with data, they were not as immediate at the time as the importance of determining stove
design factors that could reduce emissions and increase overall stove efficiency.
6-103
-------
Regional Biomass Program Impact
Page_5
• The 1988-89 Northeast Cooperative Wood Stove Study in Glens Falls, NY
After the promulgation in 1988 of the EPA regulations setting standards for emissions
for wood stoves, to be phased-in during 1988 and 1990, the CONEG Policy Research Center,
the EPA, NYSERDA, and the Canadian Combustion Research Laboratory embarked on
another round of research aimed at further quantifying the field performance of new
technology stoves. As the previous studies had demonstrated that advanced technology
stoves were not performing in the field up to the laboratory certified standards, additional
field testing was needed not only to further quantify field performance but also to determine
the factors influencing emissions, including operator practice and equipment failure. All of
the new technology stoves involved in this research were required to meet the Oregon 1988
standards, to pass a lab stress test, to have adequate draft and sizing, and to be used by
experienced wood stove operators. The stoves were considered to be 1990 EPA-certifiable.
Three catalytic stoves and two non-catalytic stoves were tested in twenty-five homes in Glens
Falls, New York, over five week-long testing periods from January - March, 1989.
The crucial finding of this study was that the average overall emissions rate was 9.4
g/h for these stoves, a 55-60% reduction from the EPA designated conventional stove rate of
213 g/h. This finding was a statistically valid (95% confidence level) indication that the
technology for the new stoves had dramatically improved over conventional stove levels.
However, there was still a large variance among the stove types in emissions levels, and two
of the three catalytic stoves did not perform well in the field. EPA lab certification levels
were within reach for at least one of the catalytic stoves, but most of the stoves tested were
found to be about twice as high in emissions in the field as in laboratory tests. One of the
catalytic stove models had serious emissions problems in four of the five stoves after only
three months of field use. The catalyst in one stove was found to be the cause of
degradation. A second catalyst failed after two seasons. Warping of the bypass systems in
one stove was found to contribute to leaks around the catalyst and elevated emissions. It
was determined that the lack of flame impingement shielding and the high internal catalyst
temperature were factors contributing to the increased emissions.
In the noncatalytic stoves, loose bypasses and some oxidation of the baffles and bypass
support mechanisms were noted; however, there was no indication that these problems had
caused emissions degradation. A computer model was generated for the noncatalytic stoves
which indicated that emissions are lowest when wood moisture is low and bum rates are
high.
This study proved that design was a crucial factor in developing high technology
stoves and in influencing emissions. For the first time in the field studies, a catalyst was
removed for further testing and a definitive finding was reached - that catalysts do fail and
that bypasses and supports must be of sufficiently heavy construction to resist warping and
oxidation caused by the high temperatures. The study noted that, for the non-catalytic
stoves, those without bypass mechanisms actually performed better, although this was not
necessarily a cause and effect situation. These findings indicated that stove design should
consider not only emissions reduction, but also other factors such as heat exchange,
sensitivity in performance to variations in burn rates, and wood moisture.
6-104
-------
Regional Biomass Program Impact
Page 6
The Glens FalJs study was the first to indicate with surety that operator practices also
contribute heavily to emissions performance. It noted that, at least in the non-catalytic
stoves, high stack temperatures and high excess air levels resulted in a net efficiency that
was not much higher than conventional stoves. While it could not be proven at this point
without additional testing, the study noted that variables such as wood moisture, burn rate,
and draft could also be important in reducing emissions and improving overall efficiency.
The study pointed to sizing and installation factors which affect draft and burn rate as well
as operator factors (wood moisture, other fuel factors).
Several important issues were raised by the Glen Falls study: 1) research activities to
date had focused on the importance of design factors in reducing emissions, but these
findings were primarily laboratory findings; in the field, it appeared that not only design,
but also operator practices were contributing to emissions levels and efficiency; 2)
laboratories had not yet been able to predict significant "stress testing" design issues; i.e.,
how stoves perform over time under field conditions (particularly in light of the catalyst
degradation and bypass problems); and 3) if operator practices were also important factors,
further testing would be needed to isolate the specific operator-controlled factors, such as
wood moisture, bum rate, and draft, that affected emissions and efficiency the most.
• The 1990 Second Season Glens Falls, NY, Study
In 1990 the Canadian government sponsored a second season follow-up study on the
Glens Falls, New York, stoves that were part of the 1988 study. This study was conducted to
measure the degree of equipment degradation after two seasons in the field. As in the
previous studies, the OMNI testing procedures using the AWES and the Data LOG'r were
used to measure field performance; additionally, the study conducted a laboratory analysis
of two failed catalysts.
Emissions in three of the four catalytic stoves (all one model) demonstrated no
increase in the second season. However, the one catalytic stove that showed elevated
emissions after one season continued to exhibit increased emissions throughout the second
season. One of the other stoves actually performed at 2g/h for both seasons, indicating the
importance of stove design and proper sizing with regard to longevity of emissions
reductions. All five of the other model catalytic stoves had elevated emissions after the
second season.
The two failed catalysts were evaluated in the laboratory and performed almost
identically to the field results. An important facet of this study was the reliability of the
"simulated real world emissions test" in the laboratory in predicting field performance. The
laboratory test was able to simulate field performance by reproducing a number of field
variables, such as wood type, wood moisture, diameter, weight, loading pattern, stack draft,
and some operator procedures (such as burn rates).
The project determined that the best performing wood stove was used at a lower burn
rate than is typical for average New York winter weather. The stove was "oversized" and
did not bum hot These findings suggested that the catalytic burning technology might be
cleanest in a warm climate where bum rates are usually lower but used in combination with
a strong draft to assure quick catalyst lightoff.
6-105
-------
Regional Biomass Program Impact
Page 7
• On-going Stress Testing
After finding that equipment degradation was a major factor contributing to emissions
elevation in cataJytic wood stoves, the Northeast Stove study group continued to sponsor
further laboratory testing to simulate the experience of wood stoves under field conditions.
Sponsored jointly by CONEG, NYSERDA, EPA, the Canadian Combustion Research
Laboratory, the Wood Heating Alliance, and the Oregon Department of Environmental
Quality, this ongoing research used OMNI Environmental Services to develop and test a
stress methodology which could simulate in days a minimum field experience of one season.
As described by OMNI, the test includes a cold start, air controls set at maximum,
keeping the loading door open for a few minutes and never completely closing it thereafter,
an opened bypass for the first ten minutes of operation, fuel reloading when stack
temperatures fall back to 500 degrees F., and a 24-hour continuous burn cycle for a period of
10 to 25 days. Early 1991 results indicated that stress tests matched closely the field
degradation findings of catalytic stoves in which degraded catalysts and metal warping
occurred in the lab conditions. For the non-catalytic stoves, oxidation of the baffle occurred.
Additional testing procedures are focusing on the causes of catalyst degradation and the
design changes that can be made to improve long-term operation. The linking of the
laboratory and field results was a significant product of this research.
• 1988-90 Field Testing in Crested Butte, Colorado
In the summer of 1988, the Town of Crested Butte, Colorado, initiated a wood stove
replacement program as an attempt to reduce particulate matter from wood stoves.
Colorado has seven areas of non-attainment with the 1987 EPA ambient air quality standards
for PM-10 (particulate matter of 10 microns or smaller). The Town of Crested Butte, a former
mining town with high wood burning usage, had adopted in 1982 a stringent ordinance
prohibiting wood stove or fireplace installation in new construction unless certain insulation
and "R-value" standards were in place. This was followed in 1986 by a voluntary stove
replacement program to provide incentives for replacement of conventional wood stoves
with newer models certified to either EPA or state of Colorado standards. With a high level
of public consciousness about the wood stove emissions problem and the replacement
program in place, Crested Butte offered a unique site for the EPA, the Wood Heating
Alliance (WHA), and the state of Colorado Department of Health to sponsor a study on
wood stove emissions before and after installation of certified model appliances.
The Town contracted the research to Dennis R. Jaasma, of the Virginia Polytechnic
Institute, to determine particulate emissions and carbon monoxide and dioxide levels
attributable to conventional and certified stoves. This field study was the first to use a new
sampling technique, developed by the Virginia Polytechnic Institute, for on-site
determination of particulate emissions. This type of sampling technique allows emissions
results of both particulate matter and CO to be obtained within 72 hours of the sampler
retrieval.4
-------
Regional Biomass Program Impact
Page 8
For this study, thirteen stoves were monitored during the first winter (1988-89), two of
which were certified catalytic models and eleven of which were conventional stoves. In the
following season (1989-90), twenty-four stoves were monitored, including seven
conventional, twelve catalytic, and five non-catalytic stoves. The certified stoves were both
EPA and Colorado certified. Wood was weighed on-site prior to use and operators were
instructed to use the "designated wood pile" during the study.
The conventional stoves were found to exhibit 47% higher particulate matter emissions
than the 15g/kg value recommended by EPA for airshed modelling. On average, the certified
stoves reduced particulate matter by 53% and CO by 49%, although there were significant
differences in performance among the various models of certified stoves. The two non-
catalytic stoves achieved levels of 71% lower particulate matter and 48% lower CO in one
model, and 44% and 19% lower in the other model. Proper sizing and adequate bum rate
were deemed crucial to the lowered emissions and efficiency of the non-catalytic stoves.
Another related finding of this study was that catalytic function over time, as in the previous
studies, was a crucial factor contributing to increased emissions in some models of catalytic
stoves; replacing the catalyst as soon as this data developed contributed to an immediate
decrease in emissions.
Research Findings Influencing the EPA "REG/NEC" Process and Final 1988 Regulations
During the EPA "reg/neg" process, only the initial findings of the first phases of
research on wood stoves were available. Given that Oregon was the only state to have
issued regulations on wood stoves, and to have some research testing data to back up their
emissions policies, the EPA was seeking input primarily on the design factors contributing to
reduced wood stove emissions. As the improved technologies were just emerging in
response to the need for deaner-buming stoves, the EPA was interested in the results from
catalytic add-ons, retrofits, and new stoves.
The first CONEG study was being conducted as the "reg/neg" process was underway.
Preliminary indications from this study can be identified as having a major influence on at
least three aspects of the final New Source Performance Standards: 1) the setting of different
standards for catalytic and non-catalytic stoves; 2) the requirement that quality assurance
testing be conducted; and 3) the consideration of bum rate and fuel cycle conditions for the
testing methodology, as well as the field testing protocol.
obtained from the LOG'r. Some interviewees noted that the VPI sampling technique, while faster
and less costly, may lose resolution of mass over time, causing it to be less accurate when
measuring emissions of 6g/hr or less.
6-107
-------
Regional Biomass Program Impact
Page?
• Equipment Auditing and Adaptation
Given that an entire series of research activities were using the same OMNI AWES
and Data LOG'r methodology and technology, the EPA was prompted to test the equipment
numerous times for quality assurance and correlation with its own Methods 5G and 5H. The
overall result of this monitoring was improved technology and increased public confidence
in the testing results. A major evolution of the equipment occurred in 1989 with the on-set
of the Glens Falls study, when an EPA audit uncovered certain deficiencies within the AWES
system. In January, 1989, the oxygen cell was discovered to be unreliable during the field
testing, and was replaced with a new cell which has proved to be extremely reliable since
then. As a result of the on-going EPA equipment auditing, functional adaptations were
made in the equipment which made the results much more reliable over time.
Over the last ten years, only one wood stove emissions study (the Crested Butte Co.
Study) has been conducted without using the OMNI equipment As a result of its
widespread effectiveness and accepted authority in this area, the OMNI technology continues
to be used in more recent stress testing (sponsored jointly by the Canadian Combustion
Laboratory, CONEG, the EPA, and the Oregon DEQ). It has also attracted the attention of
the Bonneville Power Administration in Portland, Oregon, in researching the emissions for
advanced technology small-scale biomass combustors relative to wood stoves. OMNI
continues as well to research health impacts and costs resulting from smoke in EPA non-
attainment areas, such as Oregon; Additionally, OMNI has assisted the California Air
Resources Board and the Denver Brown Qoud Study in chemical balance modelling as a
direct result of its experience with bum cycles gained from wood stove research. Finally,
OMNI's adaptation of its equipment to provide data for pellet stove and masonry fireplace
research continues to set the standard for wider use of emissions testing technology,
including the possible setting by EPA of an "AP-42H number for masonry fireplaces.
. Future Considerations
Generally, the first two rounds of research through the time of the Wood Heating
Alliance 1989 forum in Chicago, provided the wood stove industry with sustained and valid
technological data to support design and quality control efforts. While there continues to be
mixed reaction to the stress testing, there has been positive feedback from the industry
regarding the support provided by testing for the field performance of new equipment in
particular. At a March, 1992, trade show, Jim Hermann, from Earth Stove and incoming
Chairman of the Hearth Products Association, remarked that "we now have the technical
data to support the performance of new technologies in the field." Research efforts brought
the wood stove manufacturing industry into a positive relationship with government and
other consumer groups, particularly since the relationship was not just defined by regulatory
action, but also by "research which could assist the design".1
The next ventures in testing will focus on continued durability concerns, particularly in
areas designated by the EPA as "non-attainment" areas in terms of the Clean Air Act
requirements. These areas include parts of Oregon and Colorado as well as certain river
towns in New England that experience the "air inversion" quality problem. Regulatory
pressure will be sustained by the durability testing as discount rates for catalytic technology
and emissions trading by utilities lead to increased efforts to "trade" various cleaner burning
technologies.
5 Noted by Skip Hayden of the Canadian Combustion Research Laboratory.
6-108
-------
Regional Biomass Program Impact
Page 10
Additionally, the CONEG/EPA collaborative that has been advanced by the wood
stove research will lead to new venues as the field of air toxics continues to be researched.
The transferability of the wood stove testing technology, already evidenced by the masonry
fireplace and heater testing, will lead to even more collaboration efforts which may impact
future Clean Air Act statutory changes.
While the federal funding for regionally sponsored biomass program research has
remained steady over the last five years, some areas are continuing to support on-going
stress-testing as well as broadening the emissions testing to include formerly exempt heaters
such as the masonry heaters or the pellet stoves. Some states have used Oil Overcharge
funding to provide wood stove replacement loans or grants, particularly on behalf of low
income households. The EPA non-attainment areas in particular are continuing efforts to
reduce particulate emissions, and these efforts vary from the use of "smoke police" in Oregon
to detect concentrations of wood stove emissions to setting opacity limits in Klamath Falls.
The state of Colorado and several western localities have passed laws to mandate that new
homes may only accept certified stove installations.
Much of the continued testing appears to be generated in areas where non-attainment
of clean air is an issue. Oregon has six non-attainment areas, and one interviewee noted that
a 20-40% reduction in wood stove emissions resulting from equipment changeouts could
significantly impact this problem. Financing for such programs, however, is limited. Last
year, a dedicated tax on cord wood almost passed the state legislature with the backing of
large industry, the wood stove industry, and the environmental community in a unique
coalition.
With the continued use of wood for heating purposes, particularly in the non-
attainment areas, reduced emissions will remain an important goal. The 1990 census
indicated, surprisingly to a number of wood stove representatives, that the use of wood as a
primary heating source for all households increased from 17% in 1980 to 21% in 1990. Given
this increase, and the need to bring the non-attainment areas into EPA compliance, there is a
continuing need for sustained production and/or updating of cleaner-burning wood stove
equipment
Building on the findings of the later research which indicated the importance of
operator practices in reducing wood stove emissions, one of the current endeavors of the
northeast regional biomass programs will be a focus on increased consumer awareness and
education about the benefits of clean wood burning. This campaign, the "Lessons Learned"
project, will promote various public relations efforts designed to address the benefits of
wood heating in the northeast relative to other heating sources, the need to use certified
wood burning appliances, how to bum wood correctly, and how to maintain wood burning
appliances.
With increased public focus on responsible wood burning techniques and technology,
wood stove use can be viewed as a valid domestic renewable and cost-effective resource,
particularly in areas of the United States such as the northeast which continue to rely heavily
on foreign-provided fossil fuels.
6-109
-------
k
Computer/Data Logger
Thermocouple
(Room
Temperature)
Computer
ConLog Software
Exhaust Return
0>2 Input
AWES
Heated Chamber
Pump Control
Thermostat
and Heater
Og Sensor
Inlet
Thermocouples
-n
Temp.
Relay
a
0
Solenoid
Koto-
meter
Row
Rcstrictcr
Filter
XAD
Cartridge
Pump
Auxiliary
Heat
Source
Silica
Gel
Gas Bag
Time
Totalizer
Flow
Controller
Calibration
Gas Inlet
Port
t
f
Masonry Heater
Figure 1. Schematic of AWES/Data LOG'r system.
CD
s-
ro
3
V
r*»
M*
n
o
* C1
d 2
u H
rr
f
o
o
cn
C/>
OMNI
OMNI
Environmental
Services, Inc.
Not to Scale
Project No.: 80119.01
Date: February 19, 1992
Drawing: 80119-01 .DRW
-------
EXHIBIT 2
CHRONOLOGICAL LIST OF WOOD STOVE STUDIES
L 1985-86
o Northeast Cooperative Wood Stove Study (CONEG study), Vermont
and New York
o Bonneville Power Administration, Portland, Oregon
n. 1986-87
o Northeast Cooperative Wood Stove Study (CONEG study), Vermont
and New York (2nd season)
o Whitehorse Efficient Woodheat Demonstration, Whitehorse, Yukon
Territory, Canada
o Emission Sampling Systems Comparability Study, Portland, Oregon
o Catalytic Retrofit/Add-on Devices Evaluation, Portland, Oregon
"" EPA Regulations Promulgated February 26,1988
m. 1988-89
o Northeast Cooperative Wood Stove Glens Falls First Year, Glens
Falls, New York
o B.E.S.T. Project, Medford, Oregon
IV. 1989-90
o Northeast Cooperative Wood Stove Qens Falls 2nd Year, Glens
Falls, New York
o B.E.S.T. Project, Medford, Oregon (Phase 2)
o Klamath Falls Study, Klamath Falls, Oregon (Canadian Combustion
Research Laboratory and WHA)
o On-going Stress Testing, Oregon
o Crested Butte, Colorado, Colorado Dept of Health, WHA
V. 1990-92
o Crested Butte, Colorado, 2nd Season
o Pellet and Masonry Heater Testing, Western day Products Assn.
o Diversified Fuels Study, Bonneville Power Administration and
Oregon Department of Energy
o EPA Stress Test Continuation
6-111
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before comple
i. report no. 2.
EPA-600/R-94-008
3.
4. TITLE and subtitle
Proceedings: The 1992 Greenhouse Gas Emissions and
Mitigation Research Symposium
5. REPORT DATE
January 1994
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Sue Philpott, Compiler
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING OROANIZATION NAME AND ADDRESS
Acurex Environmental Corporation
4915 Prospectus Drive
Durham, North Carolina 27713
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-09-0131, Task 11-25, and
68-D2-0063 Task 17
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air and Energy Engineering Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Proceedings; August 1992
14. SPONSORING AGENCY CODE
EPA/600/13
is. supplementary notes project officer T.K. Janes is no longer with the Agency.
For details, contact Keith J. Fritsky, Mail Drop 63. 919/541-7979.
i6. abstract report documents the 1992 Greenhouse Gas Emissions and Mitigation
Research Symposium, held in Washington, DC, August 18-20, 1992. The symposium
provided a forum of exchange of technical information on global change emissions
and potential mitigation technologies. The primary objectives of the meeting were
dissemination of technical information and education in recent research. Oral papers
along with an international panel discussion, overheads, slides, and a GloED demon-
stration provided for lively exchanges in the following areas: activities in EPA,
U.S. Department of Energy (DOE), and Electric Power Research Institute (EPRI) on
greenhouse gas emissions and mitigation research, and EPA/AEERL's global emis-
sions and technology databases; international activities of selected industrialized
and developing countries; carbon dioxide (C02) emissions and their control, disposal
and reduction through conservation and energy efficiency, and carbon sequestration
including utilization of waste C02; methane (CH4) emissions and mitigation technolo-
gies including such topics as coal mines, the natural gas industry, key agricultural
sources, landfills and other waste management sites, and energy recovery by fuel
cells; biomass emission sources and sinks, including cookstove emissions and con-
trol approaches; and solar and renewable energy sources.
17. KEY WORDS AND DOCUMENT ANALYSIS
3. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. cosati Field/Group
Pollution Biomass
Gases Solar Energy
Greenhouse Effect Wood
Emission Stoves
Carbon Dioxide Earth Fills
Methane Fuel Cells
Pollution Control
Stationary Sources
Greenhouse Gases
Woodstoves
13 B 08A.06C
07D 03B
04A 11L
14G 13 A
07B 13 C
07C 10B
13. DISTRIBUTION STATEMENT
Release to Public
19 SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
534
20. SECURITY CLASS (This page)
Unclassified
22 PRICE
EPA Form 222D-1 (9-73) 6~112
------- |