United States           Office of Air and Radiation     EPA 400/1-89/002
Environmental Protection      Washington D.C. 20460      August 1989
Agency


Reducing Methane Emissions


From Livestock: Opportunities


and Issues
                                   Pruned on Recycled Paptr

-------
REDUCING METHANE EMISSIONS FROM LIVESTOCK:

         OPPORTUNITIES AND ISSUES
                   by:

             Michael  J.  Gibbs
                Lisa  Lewis
             John S.  Hoffman
   U.S.  Environmental Protection Agency
               August 1989

-------
                              ACKNOWLEDGEMENTS*
                                                                       *
    We would like to thank the many individuals who provided insights and
information that helped to produce this paper.  In particular we are
grateful to those who contributed to the workshop on Methane Emissions from
Ruminants, including:  Lee Baldwin, Don Blake, Jim Ellis, Bob Hespell, Don
Johnson, Ron Leng, Mac Safley, Henry Tyrrell, and Fat Zimmerman.

    Additionally, we are grateful to all the reviewers who helped to improve
the paper through their comments.
*  Michael Gibbs and Lisa Lewis are employed in the Universal City,
California office of ICF Incorporated.

-------
                              TABLE OF .CONTENTS

                                                                       Page

    PROLOGUE   	    1

    FINDINGS   	    1

    1.  INTRODUCTION   	    4
    1.1  SUMMARY	    *
    1.2  THE INCREASING ABUNDANCE OF ATMOSPHERIC METHANE   	    6
          CAUSES OF METHANE INCREASES 	    6
          OZONE EFFECTS IN THE STRATOSPHERE AND TROPOSPHERE	   11
          GLOBAL CLIMATE CHANGE IMPACTS  	   12
    1.3  STABILIZING THE ABUNDANCE OF ATMOSPHERIC METHANE  	   12

    2.  OVERVIEW OF METHANE FROM LIVESTOCK	   16
    2.1  SOURCES OF METHANE EMISSIONS	   16
          METHANOGENESIS WITHIN THE RUMEN 	   16.
          MANURE AS A  SOURCE OF METHANE  EMISSIONS 	   22
    2.2  GLOBAL METHANE EMISSIONS FROM RUMINANTS  	   24
    2.3  TRENDS IN SUPPLY AND DEMAND	   25
          TRENDS IN SUPPLY	   25
          TRENDS IN DEMAND	   29

    3.  STEPS. FOR REDUCING METHANE EMISSIONS FROM LIVESTOCK  	   31
    3.1  IMPROVED EMISSIONS CHARACTERIZATION   	   31
          CHARACTERIZE THE ANIMAL POPULATIONS  	   31
          DEVELOP AND EVALUATE MEASUREMENT TECHNIQUES 	   33
          PERFORM CH4 EMISSIONS MEASUREMENTS   	   35
          MODEL DEVELOPMENT:  EMISSIONS AND ANIMAL MANAGEMENT	   35
    3.2  OPTION IDENTIFICATION AND EVALUATION  	   38

    4.  EPA OFFICE OF AIR AND RADIATION PROGRAM PLAN	   45

    REFERENCES	   48


Appendix A:    Global Methane Emissions from Ruminants

Appendix B:    Impact of Reducing Livestock Methane Emissions on Global
               Warming from the Greenhouse Effect

Appendix C:    Crutzen, P.J.,  I.  Aselmann, and W.  Seiler,"Methane Production
               by Domestic Animals, Wild Ruminants,  Other Herbivorous Fauna,
               and Humans," Tellus. 1986, pp.  271-284.

Appendix D:    Workshop Agenda

Appendix E:    List of Workshop Attendees

-------
                          LIST OF EXHIBITS
                                                                   Page
EXHIBIT 1-1   GLOBAL LIVESTOCK POPULATION ESTIMATES 	    5

EXHIBIT 1-2   RECENTLY-MEASURED INCREASES IN ATMOSPHERIC METHANE   .    7

EXHIBIT 1-3   ATMOSPHERIC METHANE FOR THE PAST 1000 YEARS	    7

EXHIBIT 1-4   RELATIONSHIP AMONG CH4,  CO, OH,  AND 03   	    9

EXHIBIT 1-5   ESTIMATES OF METHANE EMISSIONS  	   10

EXHIBIT 1-6   METHANE IS AN IMPORTANT CONTRIBUTOR TO THE GREENHOUSE
              EFFECT	   13

EXHIBIT 2-1   METHANE YIELD AS A FUNCTION OF DIGESTIBILITY AND LEVEL
              OF ENERGY INTAKE	   18

EXHIBIT 2-2   ANNUAL METHANE EMISSIONS FOR COMBINATIONS OF FEED
              INTAKE AND DIGESTIBILITY LEVELS 	   20

EXHIBIT 2-3   MILK PRODUCTION PER DAIRY COW PER YEAR IN THE U.S.   .   27

EXHIBIT 3-1   STEPS FOR IMPROVING EMISSIONS ESTIMATES 	   ^2

EXHIBIT 3-2   POSSIBLE FRAMEWORK FOR A LIVESTOCK-METHANE ASSESSMENT
              MODEL	   37

EXHIBIT 3-3   PARTIAL AND PRELIMINARY IDENTIFICATION AND EVALUATION
              OF CH4 EMISSIONS REDUCTION OPTIONS   	   39

-------
                                  PROLOGUE

    A workshop entitled  "Methane Emissions From Ruminants" was held on
February  27-29, 1989  to  identify and discuss issues relating to  the role
that ruminants and other managed animals play in the global methane (CH4)
budget.   A draft version of this paper was presented and discussed at that
workshop.  This revised  version reflects the comments and discussion at the
workshop.  A  list of  the workshop attendees is presented in Appendix E.

    The following are the findings that were adopted by consensus by the
workshop  attendees.   These findings indicate that there are promising
opportunities for reducing CH4 emissions from ruminants.   Such opportunities
remain to be  assessed and demonstrated in the field.  Undertaking such
assessments and demonstrations is a recognized priority.
                                  FINDINGS

1.  Methane is increasing and will affect tropospheric air quality and
    global climate.

    1.1  The concentration of CH4 in the atmosphere is currently
         increasing at a rate of approximately 1.0 percent per year.
         This rate of increase is well characterized for the recent
         past.  Additionally, the atmospheric concentration of CH4 has
         approximately doubled in the past 200 to 300 years.

    1.2  Increasing emissions of CH4 are the primary cause of increasing
         CH4 concentrations.   Reduction in the rate of CH4  destruction
         in the atmosphere (possibly associated with increasing
         emissions of carbon monoxide) is also a factor.

    1.3  Continued increases in the concentration of CH4 in the
         atmosphere will lead to increases in tropospheric ozone
         formation (a component of smog) which is considered a threat to
         human health and the environment.  Additionally, increasing CH4
         concentrations directly and indirectly change the radiative
         properties of the atmosphere, contributing to the greenhouse
         effect.   One other hypothesis raises the concern that
         increasing CH4  concentrations could contribute to  polar
         stratospheric ozone depletion.

2.  Animals, and in particular ruminants, are an important source of CH4
    emissions on a global scale.

    2.1  Although uncertainty exists regarding the global sources of
         CH4,  it  is clear that  the major anthropogenic  sources  include:
         ruminant animals (primarily cattle, buffalo,  and sheep); animal
         wastes;  rice paddies;  biomass burning (e.g.,  forest fires);
         termites from disturbed forests (e.g.,  deforested areas);

-------
         venting and incomplete flaring of gas during oil exploration
         and extraction; leakage of natural gas during natural gas
         extraction and distribution; coal mining; and landfills.

    2.2  Current estimates indicate that anthropogenic sources account
         for about 60 percent of current CH4 emissions,  with the
         remaining emissions, from natural sources, being associated
         with swamps, marshes, lakes, and oceans.

    2.3  Animals, and in particular managed ruminant animals, produce
         significant quantities of CH* as part of  their  digestive
         processes.  Based on laboratory estimates of CH4 production by
         individual animals, and estimates of total animal populations,
         ruminant animals are estimated to account for nearly one fourth
         of the total anthropogenic emissions, or about 15 percent of
         total emissions.  These estimates do not include CH4 emissions
         from animal wastes, which may be significant.

3.  Reductions in CH4 emissions from animals will  assist in reducing the
    rate of CH4 increases, and may be one important component in attempts to
    stabilize atmospheric CH4 concentrations.

    3.1  Based on the current imbalance of CH4 emissions and destruction
         in the atmosphere, a 10 to 20 percent reduction in
         anthropogenic CH4 emissions is required to stabilize
         atmospheric concentrations at their current levels.

    3.2  Given the diversity of CH4 emissions sources, reducing
         emissions from one or two sources will not be sufficient for
         stabilizing atmospheric concentrations.  Instead, small
         reductions in each of the major sources will likely be
         preferred.

    3.3  A 50 percent reduction in CH4 emissions from ruminants will
         contribute about 50 to 75 percent of the emissions reductions
         needed to stabilize atmospheric CH4 concentrations.

    3.4  While many uncertainties exist, it would appear that with adequate
         resources and appropriate policies, emissions of CHA from livestock
         could be reduced from 25 to 75 percent with current resources
         directed toward raising livestock.  These reductions could be
         decreased if significantly larger amounts of resources were
         devoted toward growing livestock.  Reductions in CH4 emissions
         would ordinarily contribute to higher animal productivity.

4.  In order to take rational steps to reduce CH4  emissions from ruminants,
    current emissions must be better characterized and options for reducing
    emissions must be identified and evaluated.

-------
4.1  Emissions from animals in developing countries remain
     uncertain.  Techniques for measuring these emissions must be
     developed and undertaken.

4.2  Emissions from animal wastes have not been adequately
     quantified.  Animal waste management practices must be
     characterized, and emissions rates for various practices must
     be estimated.

4.3  Although potential options for reducing emissions may be
     promising, they must be demonstrated and evaluated.  Research
     for identifying long term options must be undertaken.

4.4  The scientific infrastructure exists to greatly increase levels of
     research to find solutions to limiting CH4 emissions from
     livestock.

-------
                              1.  INTRODUCTION
1.1 SUMMARY

    It is well documented that the global atmospheric abundance of methane
(CHj,) is increasing.  Recent rates of increase are on the order of one
percent per year, or about 0.017 ppmv per year (ppmv is parts per million
by volume).  This increased abundance of CH4 will have important impacts on:
the stratospheric ozone layer; background levels of tropospheric (i.e.,
ground-based) ozone; and global climate.

    Managed livestock, and in particular, ruminants, are important
contributors to the increasing abundance of atmospheric CH4, accounting for
on the order of 15 to 20 percent of annual CHA emissions.   As shown in
Exhibit 1-1, the population of managed ruminants has been increasing
globally for over 30 years.  According to FAO statistics, the numbers of
managed ruminants in 1987 included:  1.3 billion cattle; 1.2 billion sheep;
0.5 billion goats; 0.1 billion buffalo; and 19 million camels (FAO, 1988).
In recent years, animal numbers have steadied or declined in the U.S. and
Europe; however, growth continues among developing nations.

    CH4 production is  part of the  normal digestive activity of ruminants.
The primary mechanism producing CH4 in these animals is methanogenic
bacteria in the rumen converting fermentation products (primarily carbon
dioxide plus hydrogen or formate)  into CH4.   CH4  is  also produced and
emitted to various degrees from livestock wastes.  Reducing CH4 emissions
from livestock-agriculture systems alone will help to reduce the rate of
increase in the abundance of CH4.   In concert with reductions from other
sources, it could help to stabilize the CHA  abundance  in the atmosphere at
roughly current levels.

    At this time,  numerous questions and uncertainties remain regarding
sources of CH4  emissions  and options  for reducing emissions.  Two  points are
clear, however:

    o    With its relatively short atmospheric lifetime (on the order of 10
         years),  CHA is a good candidate for stabilization.   An overall
         emissions reduction on the order of 10 percent is required in
         order to prevent the CH4  abundance  from  continuing to increase (see
         WHO (1986), pp.  92).

    o    Given the diverse nature  of the sources  of CH4 emissions,
         addressing emissions from one or two sources  will not be
         sufficient to .stabilize the abundance of CH4  in the atmosphere.
         Instead,  opportunities for reducing emissions from all sources
         must be  identified and evaluated.

-------
                                 EXHIBIT 1-1

                   GLOBAL LIVESTOCK POPULATION ESTIMATES
      1400
      1300-
       1200-
       1100-
       1000-
       900-
       800-
       700
               1956
1960
1966
1970
1975
                                 1980
1966
    These data are suggestive  of the recent changes in the global population
of managed animals.  The  recent data obtained from the FAO for this exhibit
are updated versions of data previously-published by FAO in their
production yearbook series.  Although these animal population data must be
viewed as uncertain due to difficulties in counting animals (particularly  •
in developing countries),  the  data are suggestive of a trend of increasing
animal populations globally.

    The total population  of cattle,  buffalo, goats, and sheep are combined
into "Livestock Units"  using the following population multipliers: cattle -
0.8; buffalo - 1.0; sheep - 0.1, goats - 0.1.

    Sources:   Data through 1960 from:  FAO, FAO Production Yearbook. Food
               and Agriculture Organization of the United Nations, Rome,
               Italy, selected years.

               Data after 1960 from:  FAO, Livestock Numbers Database
               Printout,  Food  and Agriculture Organization of the United
               Nations, Rome,  Italy, 1988.

-------
This investigation of livestock-related emissions is only one part of the
diverse set of analyses required to identify and evaluate options for
stopping the increasing abundance of CH4 in the atmosphere.

    This paper summarizes the issues associated with reducing CH4 emissions
from livestock-agriculture systems.  For those involved in addressing
global change, it provides a framework for addressing CH4 emissions from
ruminants and as such is a foundation upon which additional data collection
and analysis can be built.  The challenge is to build upon existing data in
a manner that identifies preferred opportunities for reducing CH4 emissions
as quickly and as efficiently as possible.

    For those involved in animal sciences and animal management, this paper
identifies important gaps in understanding and data that require research
and analysis.  Additionally, it demonstrates that there are opportunities
to apply expertise and develop options for reducing emissions.  Finally, it
highlights that coordination is required among domestic and international
institutions of many types.
1.2 THE INCREASING ABUNDANCE OF ATMOSPHERIC METHANE

    CAUSES OF METHANE INCREASES

    The level of CH4 in the Earth's atmosphere is defined by emissions (from
natural and anthropogenic sources) and natural destruction processes.  It
is well documented that the atmospheric abundance of CH4 is currently
increasing (see for example:  Khali1 and Rasmussen (1986) and Blake and
Rowland (1988)).  Recent rates of increase are on the order of about 0.017
ppmv per year, or about one percent of the current average global level of
abundance of about 1.7 ppmv (see Exhibit 1-2).

    Based on analyses of air trapped in the ice sheets of Greenland and
Antarctica, long term CH4 trends have been established.   The data indicate
that the abundance of atmospheric CH4 has increased primarily in the last
300 years, with relatively stable levels prior to that time at about one-
half the current levels (see Exhibit 1-3).  This increase in abundance must
be caused by an imbalance in the sources (i.e., emissions) and sinks (i.e.,
destruction) of CH4.   The mechanisms leading to these observed increases in
CH4 are believed to be:   (1) increases in emissions;  and (2) possibly
decreases in its rate of destruction.

    The destruction of atmospheric CH4 is driven primarily by its oxidation
in the troposphere (i.e., the lower atmosphere) and stratosphere by the
hydroxyl radical, OH.  Relatively smaller amounts of atmospheric CHA are
consumed by aerobic soils.  It has been postulated that the observed
increase in atmospheric CHA may be in part due to a decline in the rate of
CH4 destruction by OH.   Such a decline could be caused by a decrease in the
abundance of OH, which could occur as the result of increasing emissions of
both CH4 and carbon monoxide (CO).   Exhibit 1-4 displays a simplified

-------
                             EXHIBIT 1-2

         RECENTLY-MEASURED INCREASES IN ATMOSPHERIC  METHANE
                      1971 79 M  «1  12  S3 M  M M «7 81
Source:    Blake,  D.R. and F.S. Rowland,  "Continuing Worldwide Increase
           in  Tropospheric Methane, 1978  to  1987,"  Science. March 4,
           1988.
                              EXHIBIT 1-3
             ATMOSPHERIC METHANE FOR THE PAST 1000 TEARS
             1800

             1600

             1400

             1200

             1000

              800

              800

              400

              200
               -1000
•800
-200
                                •600     -400
                                Tkne Before Prssflfit
Source:    Khalil,  M.A.K., and R.A. Rasmussen,  "Trends of Atmospheric
           Methane:  Past, Present, and Future," in:  Proceedings of the
           Symposium on CO; and Other Greenhouse Gases.  Brussels,
           Belgium, November 1986.

-------
                                      8

 relationship among CHA,  CO, OH,  and 03.  Although trends in the emissions
 and abundance of CO remain to be established,  large  anthropogenic sources
 of CO have been identified,  adding support to  the  assumption that CO
 emissions and abundances have increased substantially.1

     Unfortunately,  time trends for OH are  not  sufficiently well  defined in
 order to establish its relative importance in  the  measured increase in CH4.
 It is likely, however, that decreases in OH are  not  sufficient to explain
 completely the observed increases in  CH4 (Bolle, Seller and Bolin (1986),
 p. 166).  The relative importance of  changes in  emissions and OH levels
 continues to be investigated.

     Emissions estimates for CH4  remain  somewhat uncertain.  The primary
 anthropogenic sources of CH4  emissions  include:  ruminant animals  (primarily
 cattle,  buffalo,  and sheep);  rice paddies;  biomass burning (e.g.,  forest
 fires);  termites  from disturbed forests (e.g., deforested areas);  venting
 and incomplete flaring of gas during  oil exploration and extraction;
-leakage  of natural  gas during natural gas  extraction and distribution;  coal
 mining;  and landfills.  Current estimates  indicate that these sources
 account  for about 60 percent  of current CH<,  emissions, with ruminant animals
 accounting for nearly one fourth of this,  or about 15 percent of the total.
 The remaining emissions,  from natural sources, are associated with
 wetlands,  marshes,  lakes,  and oceans.

     In order for  the atmospheric abundance  of  CH< to be increasing at the
 observed rates,  it  is believed that CHA emissions must be increasing.
 Increasing populations of ruminants,  increasing  area planted  to  rice (and
 increases  in the  "double cropping" of rice;  i.e.,  planting two crops a
 year), increasing production  and distribution  of natural gas,  and
 increasing production of coal have been cited  as evidence of  increasing CH4
 emissions  from these sources.   Some of  the  key natural sources (e.g.,
 wetlands and marshes),  have reduced in  area due  to development pressures in
 various  parts of  the world.   Consequently,  increasing emissions  from
 anthropogenic sources are  believed to be the primary cause of increasing
 emissions.2

     Exhibit  1-5 presents  estimates of the  time trend of emissions  and
 estimates  of recent  emissions.   As shown in the exhibit, the  time  trend
 estimated  by Seiler  indicates  that total emissions are increasing at an
 average  annual rate  of over 1.0  percent per  year.  The estimate of current
 emissions  rates by source  shows  the relatively large uncertainties that
 remain regarding  the  allocation  of total emissions among source categories.
     1  Large anthropogenic sources include fossil fuel and biomass
combustion as well as oxidation of CH<, and other anthropogenic hydrocarbons.
See:  WHO (1986), pp. 104-106.

     2  It has also been mentioned that warming in the Arctic may be leading
to increased CH4 emissions from tundra, see WHO (1986), p. 99.

-------
                                EXHIBIT 1-4

                   RELATIONSHIP AMONG  CH*, CO, OH, AND 03
                         CH4 Emissions
    +1.0COperCH4

    -3.3 OH per CH4
    -0.7O3perCH4
    •H.OCO2perCO
    -1.0O3perCO
                          CO Emissions
+1.0COperCH4
40.5 OH per CH4
+2.703perCH4
+1.0CO2perCO
+1.0O3perCO
    Based on current models, the Implications of CH4 and carbon monoxide
(CO) emissions appear to differ for areas of low and high  nitrogen oxide
(NO) concentrations.  In low NO environments (mostly over  oceans), CH4
oxidation to CO results in a loss of OH and tropospheric ozone (03).
Oxidation of CO to carbon dioxide (C02)  further destroys 03.  In high NO
environments (mostly over continents),  CH4 oxidation to CO increases OH and
03.   CO oxidation further increases 03.

    Based on current models, the atmosphere appears to be  divided
approximately equally between the two NO environments.  Therefore,
increasing CH4 abundance will lead to reduced levels of OH and increased
levels of 03.   Additionally, CO oxidation reduces the  abundance of free OH
that can enter into reactions, effectively reducing the level of OH.
Finally, a heterogeneous CO oxidation reaction cycle (not  shown above) can
also reduce OH levels; the heterogeneous reaction cycle is not well
quantified, however.

    Source:    Crutzen, Paul J., "Role  of the Tropics  in Atmospheric
               Chemistry," in:  Dickinson, R. (ed.), Geophysiology of
               Amazonia. John Wiley & Sons:  New York, 1987.

-------
                                       10    -

                                  EXHIBIT 1-5

                        ESTIMATES OF METHANE EMISSIONS
                              (Tg of CH4 per year)*
 Time Trend  of Emissions •- Seiler:

                 aw
             I

             I
   ISO


   100


   190


fi  100


I   SO

                                                        Bnmau taming, cod mono,
                                                        •nd rutm g« Ortfcuton
                                                    l?i;$l Runwra
Current Emissions by Source -- Bolle,  Seiler, «nd Bolin:
                                                             mm: 121-177 Tg
                                                            bunfeg.aMlfflHng.
                                                        •
-------
                                     11

    Despite the various uncertainties regarding both sources and sinks, it
is clear from atmospheric measurements and ice core analyses that the
atmospheric abundance of CH4 is increasing.  Additional research and
analysis is ongoing  to improve our  assessment of the sources and sinks of
CHA, including:  atmospheric monitoring to evaluate CH4 trends  by  latitude
and CH4 isotope ratios; measurements of the emissions from key sources,
including tundra in  the Arctic, rice paddies in Asia, ruminants, and
others.

    OZONE EFFECTS IN THE STRATOSPHERE AND TROPOSPHERE

    Current atmospheric models indicate that increases in the abundance of
CH4 may increase the abundance of tropospheric and stratospheric ozone (WHO
(1986), pp. 730-732).  Increases  in tropospheric ozone, an important
component of urban "smog,"  are considered a threat to human health, crops,
forests, ecosystems  and materials (EPA (1987), p. 14-3).  The extent of
increases in tropospheric ozone due to CH4 depends on the tropospheric
abundance of nitrogen oxide (NO).   As shown above in Exhibit 1-4, CHA
oxidation leads to an increase in tropospheric ozone abundance when in the
presence of relatively high levels  of NO, as are found over the northern
continents, including North America (WHO (1986), pp. 730-731, and Crutzen
(1987), pp. 111-114).

    Based on current models, therefore, increasing levels of CH4 are
expected to worsen tropospheric ozone levels in both urban and rural
locations in the U.S.  Because the  current understanding of the link
between the levels of CH4 and tropospheric ozone is based on models of
complex tropospheric chemistry, the quantitative relationship between CHA
and tropospheric ozone remains uncertain.

   . Stratospheric ozone forms the Earth's shield against harmful solar
ultraviolet radiation.  Conventional wisdom has been that in the
stratosphere CH4 may help to protect ozone by interfering with its catalytic
destruction by chlorine and nitrogen oxides (NOX) (WHO (1986),  pp. 731-732).
Thus, it has been thought that such stratospheric ozone increases
associated with CH4 could help to offset stratospheric ozone depletion due
to chlorofluorocarbon (CFC)  emissions, thereby reducing the risks to human
health and the environment  (EPA (1987), p. 18-41).

    However, it has been suggested  recently that CH4 may in fact promote
stratospheric ozone  depletion (Blake and Rowland (1988)).  The water vapor
that is added to the stratosphere when CH4 is oxidized may provide surfaces
for heterogeneous reactions  that destroy ozone to take place.  These
reactions,' which have been  implicated in the polar ozone losses that are
currently observed (i.e., the Antarctic "Ozone Hole"), would lead to
reductions in stratospheric  ozone that outweigh the increases currently
anticipated to be associated with increasing levels of CH4.   Consequently,
the impact of rising CH4 levels  on stratospheric  ozone now is less of a
consensus.

-------
                                      12

     GLOBAL CLIMATE  CHANGE  IMPACTS

     By  changing  the radiative properties of the atmosphere, CH4 will
 contribute to  the greenhouse effect  (WMO (1986), pp.  873-874).   In  fact,  on
 a molecular basis CH4 is a more potent greenhouse gas than is carbon dioxide
 (C02) .  Of note  is  that the potency of CH4  as  a greenhouse gas may be
 amplified by associated increases  in tropospheric ozone  (because
 tropospheric ozone  is also a greenhouse gas) and stratospheric water vapor
 (see Exhibit 1-6).3  Assuming that the growth rate of global atmospheric
 concentrations of CFCs is reduced  considerably as the result of  future
 reductions in  the production and use  of these compounds,4 model calculations
 indicate  that  the CH* contribution to potential future global warming from
 human-related  emissions will be second only to C02.
 1.3 STABILIZING THE ABUNDANCE OF ATMOSPHERIC METHANE

    The current imbalance in the sources and sinks of CHA must be corrected
 in order  to stabilize CH4 concentrations at approximately current levels.
 Using estimates of current concentrations and rates of emissions  in WMO
 (1986), it is estimated that about a 7 to 14 percent reduction in current
 annual CH4 emissions is required in order to stabilize its current
 atmospheric concentration.  The range is driven by uncertainties  in the
 atmospheric lifetime of CH4.S  Because  some  emissions  are from natural (and
 presumably uncontrollable) sources, the reductions in anthropogenic sources
 required  to stabilize concentrations are about 10 to 20 percent.
     3  The amount of global warming in Exhibit 1-6 is reported in terms of
equilibrium temperature increases associated with a "direct radiative
forcing," which is a measure of the extent of the change in the radiative
properties of the Earth's atmosphere.  The actual warming of the Earth's
surface will lag behind the changed radiative properties of the atmosphere
because it takes time for the Earth to come into thermal equilibrium.  In
addition, the amount of warming anticipated will be larger than the amounts
shown in the exhibit because the direct radiative forcing is expected to be
amplified by positive feedbacks in the Earth's climate system.

     *  Significant reductions in the future use and emissions of CFCs are
anticipated as the result of the current and expected future provisions of
the "Montreal Protocol on Substances That Deplete the Ozone Layer," which
has entered into force.

     5  The higher the CH< destruction  rate  (primarily by OH),  the  shorter
its atmospheric lifetime.  The range of destruction and accumulation rates
presented in WMO (1986) implies a range of lifetimes of about 7.4 years to
15.1 years, with a middle value of about 10 years.

-------
                                     13

                                EXHIBIT 1-6

        METHANE IS AN IMPORTANT CONTRIBUTOR TO THE GREENHOUSE EFFECT
    0.10
    0.08
    CX04
    0432
Decadal Increments of Greenhouse Forcing
M»
Ml
»
-
»
•MHHM1
CO,
42
ppm
O, ttrMJ)
*L-x *^
fr
M^M

CO,
8.2
AAfll


<


••^•J*«lf*M-A
***"**2
F.>
5s;
CH4
"«**
|SCFC«


CO,
12.8
PPM


!



o,
••4
»••»••«
CFCt.
FH
^
NgO
CH,
_itr.
H«0



—•—
CO,

B.6
ppm

h




•••WMW
•tMrf
Sjg
fM
Flt
N^)
••••Mi
CH«

«
m
1
<
•

           1850 -1960
          (per decode)
I960'*
I970'«
1980's
    Decadal additions to global mean greenhouse forcing of the climate
system.  The No Feedback Warming Is the computed temperature change at
equilibrium for the estimated decadal increase in trace gas abundances, •
with no climate feedbacks included.  Based on current models, global
warming (including feedbacks) is expected to be larger, depending on the
feedbacks assumed.

    Note:      03 - tropospheric ozone
               Str. H20 - stratospheric water vapor
               Fu - CFC-11
               F   - CFC-12
    Source:  Hansen, J., et al.. "The Greenhouse Effect:  Projections of
Global Climate Change,* in:  Titus, J.G. (ed.), Effects of Changes In
Stratospheric Ozone and Global Climate. U.S. EPA and UNEP:  Washington,
D.C., August 1986, p. 206.

-------
                                     14
    Compared to other key greenhouse gases (e.g., carbon dioxide (C02) ,
nitrous oxide (N20), and CFCs), this reduction is relatively modest.  To
stabilize the concentrations of these other gases, the following reductions
are required from current anthropogenic emissions levels:  C02:  50 to 75
percent; N20:  60 to 80 percent (WHO (1986), p. 81); and CFCs:  90 to 100
percent (Hoffman and Gibbs (1988), pp. 23-27).6

    Due to the varied sources of CH4,  it is unlikely that any single
response or intervention will be capable of reducing CH4 emissions by an
amount that is sufficient to stabilize concentrations.  Instead,
undertaking options that result in modest reductions in each of the various
sources of CH4 will likely be the preferred approach for reducing overall
CH4 emissions.   Nevertheless, as one of the larger anthropogenic sources,
reductions in emissions associated with livestock could be particularly
useful.  For example, a 50 percent reduction target for livestock-related
emissions would provide on the order of half the reductions needed to
stabilize concentrations.

    Even without making other possible reductions to stabilize CH<,
concentrations, a 25 to 75 percent reduction in CH4 emissions associated
with livestock would produce a significant reduction in anticipated future
global climate change from the greenhouse effect.  As described in
Appendix B, such a reduction in livestock-related CH4 emissions could reduce
anticipated equilibrium warming increases on the order of one to six
percent by 2100 (depending on the assumptions used).

    Because no single human-related activity accounts for a large fraction
of emissions of gases that contribute to the greenhouse effect, this impact
of reducing these CH4 emissions is similar to the impacts that can be
achieved by other specific technical means, such as increasing the fuel
efficiency of automobiles, reforesting large areas, or putting emissions.
fees on the use of coal, oil, and gas (see, for example, EPA (1989)).
Additionally, unlike many of the other options for addressing global
climate change, reducing CH4  emissions from ruminants produces benefits by
improving animal productivity, particularly in developing countries.

    The technical feasibility and cost of achieving emissions reductions of
this magnitude from livestock remain to be quantified.  However, as
described below, a range of opportunities for reducing CH4 emissions from
     6  Of note is that although a 90 to 100 percent reduction in CFC
emissions will stabilize CFC concentrations over the long term, such
reductions will also stimulate the use of substitute compounds (such as
HCFC-22 and HFC-134a) which will also contribute to global warming.
Because the atmospheric lifetimes of the substitutes are much shorter than
the lifetimes of the CFCs, the anticipated impacts of the substitutes on
global climate will be much smaller than what the impacts of the CFCs would
have been.

-------
                                     15

livestock have been identified, indicating that some reductions, and
possibly large reductions may be feasible.  A major goal of this
assessment, therefore, is to identify options for achieving such a
reduction in livestock-related emissions.  These options could subsequently
be evaluated.

    Such an evaluation would include:

    o    Improved Characterization of Emissions.  Emissions rates from
         various animals and animal management practices (particularly in
         developing countries) must be better characterized.  Improved
         emissions estimates will enable emission reduction options to be
         better evaluated.

    o    Evaluation of Impacts.  Options for reducing emissions must be
         evaluated in terms of:  impact on emissions; cost; ease of
         implementation; and impact on agricultural productivity.  Given
         the obvious importance of managed ruminants throughout, the world,
         and their unique ability to convert low quality inputs (e.g.,
         agriculture byproducts that cannot be digested by humans) into
         useful products, potential technical and management options for
         reducing CH4 emissions must be evaluated within the context of the
         overall agricultural, social, and economic systems in which they
         would be implemented.

    o    Assessment of Methods of Implementation.  Institti*"ions and
         organizations involved with agriculture and agriculture products-
         must be involved in implementing emissions reduction options.

    Similar analyses are required for the other sources of CH4 emissions.
For example, improved estimates of CH4 emissions are required in order to
evaluate whether additional gas recovery during coal mining is warranted.
Continued improvements in our understanding of the global sources and sinks
of CH4 are required in order to have confidence that the levels of emissions
reductions achieved over time are sufficient for stabilizing atmospheric
concentrations.

-------
                    2.   OVERVIEW OF METHANE  FROM LIVESTOCK
 2.1  SOURCES OF METHANE  EMISSIONS

     Among  livestock,  the  ruminants  (i.e., cattle, buffalo, sheep, goats,  and
 camels)  are the  major emitters  of CHA.  The rumen, a large "fore-stomach,"
 is the unique physiological  characteristic of  ruminants that provides  the
 opportunity for  CH4 to be created within the animal.  Within the rumen, over
 200  species and  strains of microorganisms have been identified, although  a
 smaller  number  (10  to 20  species) are believed to play an  important  role  in
 digestion  (Baldwin  and  Allison  (1983), p. 462).  Rumen methanogenie
 bacteria are the source of CH4 produced within ruminants.

    'To date, methanogenesis  within  the rumen of ruminants has been the
 primary  focus of discussions regarding CH4 production related to livestock.
 However, ruminants  are  generally managed as part of an overall system,
 which may  have other  parts that also contribute to CH4 emissions.  The
 principal  system component identified that may contribute  the CHA emissions
 is the disposal  of  manure.   Significant CH4 emissions have been measured
 from lagoons used to  dispose of animal wastes, including non-ruminants,
 such as  swine and poultry (Safley and Westerman (1988)).

     METHANOGENESIS WITHIN THE RUMEN

     Rumen methanogenic bacteria are generally  a very small fraction  of the
 total population of microorganisms  in the rumen.  Although they can  convert
 acetate  (a fermentation product produced in the rumen) to CH4 and C02,  this
 pathway  for CH4 production in the rumen is believed to be of minor
 importance in animals fed adequate and balanced diets (Baldwin and Allison
 (1983), p. 469).  Instead, the conversion of hydrogen (H2)  or formate and
 C02  (produced by fermentative bacteria)  is believed to be  the primary
 mechanism by which methanogenic bacteria produce CH4 in ruminants.

     The creation of CH4  in the rumen represents energy which is  subsequently
 not  available to  the host animal for maintenance or growth.  Methods of
 reducing methanogenesis in ruminants have been investigated as part  of the
 overall attempt to improve the efficiency of rumen metabolism.  However,
 methanogenic bacteria play an important role in the complex ecology  of the
 rumen, so that simply eliminating or suppressing the activity of
 methanogens in the rumen will not "free up" energy that can be used by the
 animal.

    The extent of methanogenesi.s in individual ruminants has been .estimated
by various authors.   The rate of methanogenesis can be described in  terms
 of a "methane yield," which  is defined as the amount of CH4 produced as a

-------
                                     17

percentage of the gross food energy intake of the animal.  Most CH4  yield
estimates for ruminants are in the range of A to 9 percent.7

    Using an estimate of the CH* yield for an animal,  its total annual  CH4
emissions can be estimated by multiplying its annual gross energy intake by
the appropriate percentage (e.g., six percent) and then converting the
energy value (e.g., in megajoules, or HJ) to a mass basis (i.e.,
kilograms).  For example, if a cow consumes 60,000 MJ per year, and has a
CH4 yield of 6 percent,  the total CH4  emissions  for  the animal  would be  the
equivalent of about 3,600 MJ, or about 65 kilograms.

    Methanogenesis occurs within individual animals. and various factors
affect the rate of methanogenesis on the individual animal level.  In '
addition, however, the total amount of CH4 produced by a population  of
animals can be evaluated in terms of the amount of CH4 emitted relative to
the amount of useful product produced by the population.  The manner in
which the population is managed will influence the size of the population
required in order to produce the useful products desired, and consequently
the total level of CH* emissions.  Factors affecting methanogenesis  in
individual animals and populations are described in turn.

    Individual Animals

    The level and type of diet a ruminant consumes has a strong influence on
an animal's CH4 yield and on the amount of CH4 produced by the  animal.
Blaxter and Clapperton (1965) examined the results ->f 391 experiments on
sheep and found that the CH4 yield is primarily a function of two factors:
(1) the digestibility of the feed; and (2) the level of energy consumed in
relation to the maintenance energy requirements of the animal.  Exhibit 2-1
shows Blaxter and Clapperton's CH4 yield estimates for a range of values for
these factors.8

    As shown in Exhibit 2-1, animals with low intake levels (e.g.,  1.0- times'
maintenance energy requirements) eating low digestibility feed (e.g., 50 to
60 percent) have CH4 yields in the 6.8 to 7.4 percent range.   Such
conditions may be anticipated among animals in many developing countries.
Alternatively, animals with high intakes (e.g., 3.0 times maintenance)
eating highly digestible feed (e.g., 70 to 80 percent) have CH4 yields in
the 5.4 to 5.8 percent range.  Such conditions may be anticipated among
highly productive dairy cows in the U.S., for example.
     7  See for example:  Blaxter and Clapperton (1965) and Moe and Tyrrell
(1979) and Rumpler, Johnson, and Bates (1986).

     8  Care must be exercised in using the results presented in Exhibit
2-1.  One would not expect to find animals consuming highly digestible feed
(e.g., 70 percent or higher) at levels near maintenance.  Similarly, one
would not expect to find animals consuming low digestible feed  (e.g., 50 to
60 percent) at 3.0 times maintenance.

-------
                                     18

                                 EXHIBIT 2-1

                       METHANE YIELD AS A FUNCTION OF
                  DIGESTIBILITY AND LEVEL OF ENERGY INTAKE
                                  (percent)
                              Level of  Intake

                       1.0         1.5         2.0         2.5         3.0

     Digestibility

           50

           60

           70

           80

           90
6.8
7.4
8.0
8.6
9.2
6.7
7.1
7.4
7.8
8.2
6.6
6.8
6.9
7.0
7.1
6.6
6.4
6.3 '
6.2
6.1
6.5
6.1
5.8
5.4
5.0
     Estimates based on an analysis of 391 experiments on sheep.  Methane
yield is the percent of feed energy that is converted to CH4.   Level of
intake is measured as the ratio of total gross energy intake to total gross
energy intake needed to meet maintenance energy requirements (e.g.,  total
gross energy intake equals two times maintenance requirements).
Digestibility in percent.
     Source:  Blaxter, K.L. and J.L. Clapperton, "Prediction of the Amount
of Methane Produced by Ruminants," British Journal of Nutrition. Vol. 19,
1965, pp. 511-522.

-------
                                     19
    These CH4 yield estimates can be used to estimate expected CH4 emissions
per year by multiplying the rates by the associated levels of intake.
Exhibit 2-2 displays these estimates for a 500 kg beef steer.  The
emissions rates reflect both the CH« yield and the influence that feed
digestibility has on required energy intake.  For example, at 2.0 times
maintenance, the CH4 yield (as a percentage of total feed energy intake)
increases with increasing feed digestibility  (see Exhibit 2-1).  However,
CH4 emissions (e.g., in kilograms per year) decline (see Exhibit 2-2)
because the increasing digestibility reduces  the energy intake required in
order to meet the 2.0 times maintenance level.8

    Others have also examined CH4 emissions rates.  For example, Moe and
Tyrrell (1979) report on a total of 404 total energy balance trials on
Holstein dairy cows.  Unlike the Blaxter and  Clapperton model, Moe and
Tyrrell estimate CHA emissions as a function of the amounts of the following
feed constituents consumed:  soluble residue, hemicellulose, and cellulose.
Feeds vary in their content of these constituents.  CH« production was found
not to be correlated with the intake of other feed characteristics such as
crude protein, ether extract, and lignin.

    Using concentrate and hay feeds typically fed to U.S. dairy cows, and
the feed characteristics reported in NRC (1988), the Blaxter and Clapperton
CH4 estimates were compared to the Moe and Tyrrell estimates.  For the
chosen diet,10 the Moe  and Tyrrell  equation results  in an estimate of about
120 kg per year of CH4 emissions, for an implied CH4 yield of about 5.7
percent.  The analogous estimate based on Blaxter and Clapperton is 135 kg
per year (for a 600 kg dairy cow), or about 12.5 percent larger than the
Moe and Tyrrell estimate.  Given that the two methods for estimating
emissions are based on different data (in fact, different types of
animals), such similar estimates are in reasonably close agreement.

    Most ruminants in the world are in developing countries where levels of
intake and feed characteristics are very different from the diet of U.S.
dairy cows.  In general, ruminants in developing countries eat agricultural
     9  As with Exhibit 2-1, the results in Exhibit 2-2 must be examined
with care.  One would not expect to find animals consuming highly
digestible feed at levels near maintenance or animals consuming low
digestible feed at 3.0 times maintenance.

     10   Twenty pounds .of concentrate  per day containing barley, hominy,
rolled corn, beet pulp, millrun wheat, and cotton seeds, and fifteen pounds
of alfalfa hay per day.  This diet is estimated to provide about 2.3 times
the maintenance energy requirement for a 600 kg dairy cow, with a
digestibility of about 65 percent.  The diet is also estimated to contain
about:  4 kg/day of soluble residue; 2 kg/day of hemicellulose; and 3.3
kg/day of cellulose.

-------
                                     20

                                 EXHIBIT 2-2

                  ANNUAL METHANE EMISSIONS FOR COMBINATIONS
                   OF FEED INTAKE AND DIGESTIBILITY LEVELS
                       (kilograms per year per animal)
   Digestibility

         50

         60

         70

         80

         90
1.0



57.

52.

 NA

 NA

 NA
                               Level of Intake

                                 1.5         2.0
97.

85.

77.

 NA

 NA
 NA

117.

102.

 91.

 82.
2.5



 NA

 NA

123.

105.

 92.
3.0



 NA

 NA

 NA

114.

 94.
    Estimates are based on CH4 yield estimates from Blaxter and Clapperton
(1965) and feed characteristics from NRC (1984).   Estimates are for a 500
kilogram animal with feed energy maintenance requirements similar to that
described in NRC (1984) for U.S. beef cattle.  Energy requirements vary by
breed, sex, and climate.  Smaller animals (such as those found in many
parts of developing countries) would have lower energy requirements, and
consequently lower CH4 emissions for any given combination of level of
intake and digestibility.

    NA - Not Applicable.  One would not normally expect to observe animals
at these combinations of feed intake levels and feed digestibility.

-------
                                     21

byproducts with low digestibility (e.g., wheat straw,  rice straw,  and sugar
cane tops), with levels of intake near maintenance requirements.   For these
animals on relatively poor diets, Preston and Leng (1987) indicate that CH4
yields may be in the 9 to 12 percent range, with the level influenced by
the adequacy of ammonia levels in the rumen (Preston and Leng (1987), p.
40).  In their assessment, (based on stoichiometric considerations)
inadequate ammonia levels are related to inefficient rumen fermentation and
increased CH4 production.   If this is the case,  supplementing diets of these
animals with urea or poultry litter (good sources of rumen ammonia) could
increase rumen fermentation efficiency and reduce CH4  emissions.   Such
supplements are starting to be provided to milk cows in India, for example,
as a means of increasing milk production.
                                                                  /
    Considerable uncertainty remains regarding the levels of CH4  emissions
anticipated from animals in developing countries.  The Preston and Leng
assessments indicate larger CH4 yields than would be indicated by Blaxter
and Clapperton.  Because feed costs account for a significant portion of
the costs of producing milk and beef in developed countries, considerable
data have been collected on the performance of feeds in these areas.
However, less data are available to describe feed characteristics and
performance in most developing country situations.

    Animal Populations

    The manner in which an animal population is managed will influence the
overall level of CH4 emissions.  In general, a population of managed animals
is maintained in order to produce some set of useful products, such as
milk, meat, wool, and (in the case of animals in developing countries)
work.  The size of the population required in order to produce the desired
level of products will depend on the productivity of the animal population.

    The productivity of dairy cows is primarily measured by the amount of
milk produced per cow per year.  Significant increases in productivity have
been achieved in the U.S. and around the world due to selected breeding and
improvements in animal management.  By  increasing the amount of milk
produced per cow, the feed requirements per amount of milk produced have
declined.  Concurrent with reductions in feed requirements should be
reductions in the amount of CH4 generated per amount of milk produced.  This
result is anticipated assuming that CH4 yields per amount of feed energy
consumed remain unchanged.  The possibility that CH4 yields of more
productive dairy cows are in fact lower than average  (possibly contributing
to  the increased productivity of the cows) remains to be demonstrated
conclus ively.

    In addition to examining the productivity of dairy cattle on  a per  cow
basis, the productivity of the population  as a whole can be assessed.  At
any point the dairy cattle population includes not only lactating dairy
cows (i.e., those actually giving milk) but also  those between lactation
cycles and those that are growing (i.e., replacement heifers  that have  yet
to  give milk).  By reducing the  time between lactation cycles, increasing

-------
                                     22

the rate of maturity of replacement heifers, reducing losses due to
disease, and increasing the success rate of replacement heifers (some
replacements are not productive), the total population of dairy cows and
heifers can be reduced while still maintaining the same level of milk
production.  Along with such reductions in the population size would come
reductions in the generation of CH4.

    Similar analyses of beef cattle populations are applicable.  The primary
measures of productivity of beef cattle are rate of weight gain and feed
efficiency.  Although beef cattle breeding has not been as successful as
dairy cattle breeding in improving productivity, hormone-based growth
stimulants have been developed that increase feed efficiency about 5 to 10
percent during finishing.  These implants are used widely in the U.S. and
other countries, although they have been banned in the EEC.

    As with dairy cows, the total size of the beef cattle population can be
reduced by:  reducing losses from disease (antibiotics are currently used);
increasing the birth rate and decreasing the inter-calving interval of cows
used to produce calves for meat production (current birth rates are about
75 calves per 100 producing cows per year in the U.S., and possibly lower
in other countries); and increasing the success rate of replacement
heifers.

    In developing nations,  cattle,  buffalo,  horses, donkeys,  mules, and
camels are used as draft animals, and cattle, buffalo, camels, goats, and
sheep are used for meat, milk, fertilizer, fuel, wool, and hide production.
In some cases, the animals also form the basis for storing wealth in many
economies that are not based on a "cash" currency.  The successful
implementation of realistic methods of increasing the productivity of these
animals within the context in which they are managed remains a challenge.

    MANURE AS A SOURCE OF METHANE EMISSIONS

    The amount of CH4 generated and emitted  from manure  depends on the
manner in which the manure is handled.  If aerobic conditions exist (i.e.,
if the manure is in contact with oxygen) then CH4 generation  should be
minimal.  If the manure is maintained under anaerobic or anoxic conditions
(i.e., in the absence of oxygen), then some portion of the organic matter
in the manure may be converted to CH4.

    In many locations, manure is used as fertilizer.  If manure is spread on
dry soils, and if it decomposes aerobically, it may produce little or no
CH4.   However,  spreading manure as  fertilizer on anoxic  soils (e.g.,  flooded
rice paddies) will likely produce CH4,  possibly in large amounts.   Good
field measurements are lacking in this area and need to be undertaken.

    In addition to fertilizer, manure is used as an energy source.  The
manure may be gathered and dried, and subsequently burned.  This method of
handling manure should produce very little CH4 as the organic material is
oxidized directly to C02.   Alternatively,  the manure may be collected and

-------
                                      23

 used in a biogas (i.e.,  CH4)  generator.   In  this case  the organic material
 in the  manure is deliberately converted to CH4, which  is subsequently
 collected and used as fuel.   The use of biogas generators would result in
 CH4  emissions  only to the  extent that CHA leaked from containment or was
"incompletely burned.

     In  locations where large numbers of animals are managed in a confined
 location (e.g.,  dairies in the U.S.  and Europe, feedlots  in the U.S.,  swine
 and poultry farms in the U.S.) manure is a waste product  that requires
 proper  handling and disposal.  The waste manure may be piled up until  it
 can be  hauled away, or it may be washed into ponds.  In either case,
 anaerobic conditions are likely to exist, which would allow a portion  of
 the manure to be converted into CH4  and emitted.

     The extent of CH4 emissions from piles of manure remains  to be
 quantified.  Such emissions may be limited by the acids that are produced
 in the  wastes during anaerobic decomposition, however. CH4 emissions  from
 waste lagoons and digesters have been measured, and were  found to be on the
 order of 0.14 to 1.0 cubic meter of CH<, per  kilogram of volatile solids
 added to the lagoon,  although in at least one case, emissions were much
 higher  (Safley and Westerman (1988), p. 187, Hill (1984)).   This rate  of CH<
 emissions translates into about 12 to 85 cubic meters  of  CH<,  per metric  ton
 of manure (wet weight) added to the lagoon,  or about 8 to 55 kilograms of
 CH4  per ton of manure.   Because a 600 kilogram dairy cow produces about  15
 tons of manure per year (wet weight, Ensminger (1983), p.  737), the
 disposal of manure wastes, in an anaerobic lagoon could result in about 120
 to 825  kilograms of CH4  emissions per head per year.   This  leval of
 emissions is on the order of one to 10 times the level of CH4 emissions
 originating in the dairy cow's rumen.  Additionally, swine  and poultry
 wastes  produce similar amounts of CH4 emissions per ton of  waste,  although
 CH4  emissions directly from these animals' digestive systems  is relatively
 small.

     Although the potential for CH4 emissions from  waste .manure is large, "it
 is unlikely that this potential is fully realized.  Most  manure is either
 spread as fertilizer, burned (a common practice in developing countries) or
 allowed to remain in pastures or on ranges.   The CH4 emissions resulting
 from these practices remain to be quantified.11  Manure is  probably only
 piled up and/or washed into ponds at intensively managed  dairy farms and
 feed lots where there is no market for manure fertilizer.   Even if these
 animals are a small fraction of the total animals, the emissions from  these
 wastes  would be significant.  For example,  if 10 percent  of the manure
 produced by cattle were to decompose anaerobically, and if these wastes
 were'converted to CH4 at a rate of 0.14 m3 per  ton  of volatile  solids  added
 (a lower bound conversion rate) then on the  order of 15 Tg per year of CHA
      11  Most manure in pastures and on ranges may decompose aerobically.
 However,  at least one  researcher has detected CH4 emissions  from this  type
 of manure (Goreau and  Mello (1985)).

-------
                                      24

 (1  Tg  -  1012 g) would be produced.  This amounts to 20 percent of the
 estimate of CH4 emissions from livestock.
 2.2 GLOBAL METHANE  EMISSIONS FROM RUMINANTS

    Crutzen  et  al.  (1986) have performed the most comprehensive assessment
 of CH4 emissions from ruminants to date.  Based on a review of CH4 yields
 and feed  characteristics and consumption, Crutzen et al. estimate  the
 following:

    o     Cattle in  developed countries. Brazil and Argentina:  average
          annual CH4 emissions per head of 55 kg/yr; total annual emissions
          of  31.5 Tg/yr;

    o     Cattle in  developing countries:  average annual CH4 emissions per
          head of 35 kg/yr; total annual emissions of 22.8 Tg/yr;

    o     Sheep  in developed countries:  average annual CH4 emissions per
          head of 8  kg/yr; total annual emissions of 3.2 Tg/yr;

    o     Sheep  in developing countries and Australia:  average annual CH4
          emissions  per head of 5 kg/yr; total annual emissions of
          3.7 Tg/yr;

    o     Buffalo (virtually all are in developing countries):  average
          annual  CH4 emissions per head of 50 kg/yr;  total annual emissions
          of 6.2  Tg/yr; and

    o     Goats  (virtually all are in developing countries):  average annual
          CH4 emissions per head of 5 kg/yr;  total annual emissions of 2.4
          Tg/yr;  and

    o     Camels  (virtually all are in developing countries'):  average
          annual CH4 emissions per head of 58 kg/yr;  total annual emissions
          of 1.0 Tg/yr.

The total of these  estimated emissions is about 71 Tg/yr, over 75 percent
of which  is associated with cattle.

    In addition, Crutzen et al.  estimate an additional 2.9 Tg/yr of CH*
emissions associated with the digestive systems of pigs, horses, mules, and
humans.12  Approximately 2 to 6 Tg/yr of emissions is estimated for wild
ruminants throughout the world,  and emissions from large non-ruminants are
expected to be small.   The overall total for these sources of CH4  is  on the
order of about 80 Tg/yr.
     12  Note that these estimates do not include potential CH4 emissions
associated with wastes.

-------
                                     25
    As noted by Crutzen et al., various uncertainties remain in these
estimates.  Chief among them are realistic emissions rates for animals in
developing countries.  Appendix A summarizes various uncertainties in  the
estimates presented, and Appendix C presents a copy of the paper published
by Crutzen et al.
2.3 TRENDS IN SUPPLY AND DEMAND

    Livestock provide a diverse set of products for human consumption.
Throughout the world, livestock management, processing, and product
distribution are significant economic activities, providing important
sources of nourishment to all people.

    The current markets for livestock products reflect the complex
interaction of a diverse set of factors including:  genetic characteristics
of animals; management practices; the availability of natural,
technological, and human resources; consumer preferences; and cultural
preferences and traditions.  Government policies are used throughout the
world to influence these various factors.

    As a result of these numerous factors, the markets for livestock
products vary considerably throughout the world and are constantly
changing.  These changes will influence the levels of CH4 emissions from
livestock, as well as the opportunities for reducing emissions.

    TRENDS IN SUPPLY

    Within developed nations, the technologies and methods for intensively
managing ruminants to produce milk and meat products have increased
productivity dramatically over the past 40 years.  In particular, the
productivity of milk cows has increased substantially during this time.
Although feed intake per cow has increased with milk production per cow,
the increases in productivity have enabled larger quantities of milk to be
produced per amount of feed (and other) inputs.  The result is that larger
amounts of milk are produced per amount of CH4 emitted.  This trend may be
an important factor influencing future CH4 emissions from ruminants.

    Several of the key trends in the supply of milk and meat products from
intensively managed cattle are as follows:

    o    Increased productivity of dairy cows through selective breeding
         and improved management.  With the development of artificial
         insemination techniques and methods of evaluating bulls and cows,
         the systematic breeding of dairy animals has been performed in the
         U.S. and elsewhere with considerable success.  The high genetic
         potential of Holstein dairy cows to produce milk has been
         distributed throughout the dairy herd, and management methods have
         improved (e.g., in the preparation of balanced rations) so that

-------
                                     26

         the value of the genetic potential has been realized in terms of
         "on-the-farm" performance.  Exhibit 2-3 displays estimates of milk
         produced per cow in the U.S. since 1950.

    o    Increased reproductive efficiency of beef brood cows.  In the U.S.
         (unlike in Europe) dairy cows are not the primary source of calves
         that are grown into beef.13   Instead,  a.  separate set  of cows,
         usually referred to as "brood" cows, produce calves that are grown
         into beef.  The productivity of these brood cows are measured in
         terms of the number of offspring produced (reproductive
         efficiency) and their meat characteristics (such as weaning weight
         and carcass quality).  The reproductive efficiency of brood cows
         in the U.S. is about 0.75 calves per cow per year.  Increased
         reproductive efficiency reduces the size of the brood herd needed
         to produce a given amount of meat, and hence reduces CH4  emissions
         per amount of meat produced.

    o    Increased feed efficiency of beef animals.  The cost of feed is a
         major cost of growing meat.   Consequently, improvements in weight
         gain per amount of feed consumed are desireable (i.e.,
         improvements in feed efficiency).  Two technologies that have been
         adopted to promote feed efficiency are ionophore feed additives
         and steroid implants.  Each of these techniques increases feed
         efficiency about 5 to 10 percent (Ensminger (1987), p. 859).
         lonophores are used widely in the U.S. among finishing cattle
         (i.e., in feedlots), and among some growing cattle in pasture
         situations.  Implants are used in growing range,  pasture, and
         feedlot animals throughout the world.  Implants have recently"been
         banned in the EEC, however.

    These various advances in animal management,  combined with a relatively
saturated demand for animal products in developed nations has lead to the
stabilization or decline of animal numbers in the U.S., Europe, and other
developed nations, while milk and meat production continue to increase.
Declining animal numbers in these areas likely indicates reduced CH4
emissions from this source, although changes in animal management
characteristics (such as level and type of feed consumed and waste
management practices used) must also be considered.
     13   In Europe, most of  the beef  (70 to nearly 100 percent, depending  on
the country) is derived from the dairy cow population through the calves
produced by these cows and through the slaughter of older cows.   As such,
the cows in Europe are referred to as "dual purpose," i.e.,  they supply
both milk and meat.  In the U.S., most of the beef is derived from calves
produced by beef brood cows.  These brood cows are separate  and distinct
from the population of cows used to produce milk.  As such,  the U.S. cows
are referred to as "single purpose," i.e., they produce either milk or
meat.

-------
                                      27

                                 EXHIBIT 2-3

             MILK PRODUCTION PER DAIRY COW PER TEAR IN THE U.S.
•e-
I

CO
D
      14000
      12000  -
      10000  -
       8000  -
       6000  -
       4000  -
       2000  -
           1950    1955    1960    1965    1970    1975    1980    1985
    Source:    NMPF,  Dairy Producer Highlights. National Milk Producer
               Federation,  Arlington, Virginia, 1987, p. 5.

-------
                                      28
     These trends  may be  anticipated to  continue  into  the  foreseeable future.
 Selective breeding continues to produce productivity  gains,  and techniques
 (such as embryo transplantation) are being developed  that could improve the
 rate of improvement.   New methods of improving animal productivity (such as
 the use of synthetic bovine growth hormone,  BGH)  hold the promise  of making
 significant improvements.   Cows injected with  BGH produce 10 to 25 percent
 more milk than cows injected with a placebo (Fallert  et al.  (1987) and Mix
 (1987)).  If the  use of  BGH is  approved,  the population of dairy cows
 required to produce a given amount of milk will  decline.  Whether  this
 decline is realized,  for example in the U.S.,  depends on  how the pricing
 and subsidy system for milk changes in  response  to (or in anticipation of)
 the use of BGH (Fallert  et at.  (1987)).   Whether the  use  of  BGH reduces CH<
 emissions (as opposed to just animal numbers)  also remains to be examined,
 because feeding practices will  also likely change with the use  of  BGH.

     While methods of intensively managing livestock have  improved
 productivity in developed countries,  the productivity of  "extensive" animal
 management systems (such as village conditions in developing countries and
 range animals), has not  changed significantly  in recent years.   Many,  if
 not most,  animals in developing countries continue to be  managed in small
 herds (1,  2,  or 3 head per household) with traditional practices.   Animal
 nutrition and health remain well below  the levels seen in developed
 countries,  and hence  productivity (measured by intensive  management
 standards;  i.e.,  in terms  of the amount  of milk  and/or meat  produced per
 animal)  remains low.   Consequently,  increases  in  the  demand  for animal
 products in some  developing nations have  been  met through increases in
 animal numbers.   These increases probably imply  increased CH4 emissions  from
 this source.

     In many developing countries  animal  feed resources are constrained.  In
 such situations,  animal  numbers  in the countries  cannot increase,  even as
 human populations have increased.   The growing demands for animal  products
 (mostly milk products) in  these  countries  are  partially met  through imports
 (which are  limited by available  economic  resources),  and  largely go unmet.

     Development projects undertaken by the World  Bank and others have among
 their objectives  improvements in animal productivity  in developing
 countries.   For example, the  largest ongoing dairy development project  is
 "Operation  Flood"  in  India.   Funded through  the sale of EEC-donated dairy
 products and World Bank  loans, this program has among its obj ectives  to:
 improve  transportation links between rural milk producing areas  and urban
 population  centers; improve storage and processing facilities; establish
 marketing cooperatives;  develop cattle feed plants; and improve  genetic
 characteristics of dairy cattle.

    To the extent  that projects such as Operation Flood are successful  in
 increasing animal productivity,  they likely reduce the rate of CH4  emissions
per amount of product  
-------
                                     29
    TRENDS IN DEMAND

    The demand for milk and meat products are the primary factors driving
the sizes of animal populations in developed countries.   In developing
countries, the needs for draft power and a mechanism for storing wealth are
also important factors.

    The per capita consumption of animal products in most developed nations
has stabilized in the past 10 to 20 years in terms of calories, protein,
and calcium consumed (FAO (1986)).  In some areas, for example the U.S.,
per capita poultry consumption has increased while per capita red meat
consumption has remained relatively constant (American Meat Institute
(1987)).

    Future trends in demand for animal products in developed nations will
likely be driven by changes in the size and age structure of the population
and by changes in consumer preferences.  For example, demand for low-
cholesterol and low-fat products are leading to developments of low-fat
meat and dairy products (Dunkley (1982)).  Changes in the complex demand
for milk products (currently driven primarily by the demand for milk fat
for use in milk products like cheese and ice cream) could influence
breeding and feeding strategies, possibly affecting CH4 emissions as well.

    Demand for low-fat meat could influence the feeding strategies and
slaughter ages of feed-lot-finished cattle.  Similarly, concerns over
antibiotics  (used in therapeutic treatments of animals in feedlots) could
also influence feedlot finishing practices.  Shifts to range- and pasture-
fed beef  (if they occur) could increase feed requirements (e.g., ionophore
feed supplements are more difficult to administer to range-fed animals),
possibly  increasing total CH4 emissions.

    Consumer concerns over hormone use could also affect the manner .in which
meat and milk are produced, and hence, CH4 emissions.  The recent EEC ban on
the use of steroid implants is an example.  Such concerns could also affect
the potential use of bovine growth hormone and the realization of its
benefits  in  terms of CH4 emissions reductions (see above).

    Increasing population and income among developing countries point to
increasing demand for animal products  in these areas.  In general, .stage  of
development has been found to be a good indicator of food consumption
patterns  (Rojko et al.  (1978)).  Consequently, as development proceeds  and
as  incomes continue to rise in developing countries, meat and dairy product
production and consumption may be anticipated to  increase as well.  The
extent  of these increases depends not  only on rates of development, but on
regional  and local tastes and customs  as well.

    Given that extensive animal management systems predominate  in developing
countries, increasing  demands for meat and milk products are expected to  be
translated into increasing populations of animals.   Increases  in animal

-------
                                     30

productivity could reduce the rate of animal population growth, and the
possibility of such increases being large enough to allow consumption to
increase while holding animal populations constant remains to be
investigated.

    The demand for draft power also has an important influence on animal
populations in developing nations.  Virtually all buffalo are in developing
nations, and many of these are used for draft power.  An increasing human
population will likely increase the need for draft power, while
mechanization could reduce the demand for animal draft power.  However, the
costs of equipment, maintenance, and fuel limit the potential impact that
mechanization will have in the short term.

    The combined influence of these various trends in supply and demand on
CH4 emissions is unclear.   Although animal numbers in developed nations are
not growing and productivity is continuing to increase,  animal management
practices may be leading to increased energy consumption per animal and
increased disposal of animal wastes in lagoons.  The net effect on CH4
emissions from these animals is therefore ambiguous and remains to be
assessed.  Animal numbers and feed consumption are increasing in developing
nations, leading to increased CH* emissions.   The rate of increase,  however,
remains to be assessed.

    In its recent analysis of options for stabilizing global climate, EPA
(1989) estimated future CH4 emissions from livestock under a variety of
assumptions.  Assuming that policies are not pursued to  reduce these
emissions, EPA estimated that CH4 emissions from animals could increase by a
factor of about 2.4 by 2100.  In light of anticipated population growth (by
a factor of about 2.0 to 2.6 by 2100) and economic growth (historically,
the consumption of animal products per capita (meat and milk) have risen
along with income per capita), such emissions growth seems plausible.
However, such emissions growth will not be possible if important resource
constraints prevent the supply of animal products from increasing along
with population and income.  Such constraints have not been proposed to
date.

-------
           3.   STEPS FOR REDUCING METHANE EMISSIONS FROM LIVESTOCK
    To reduce CH* emissions from livestock, systematic investigations are
required in two main areas:  (1) emissions  characterization; and (2) option
identification, evaluation,  and implementation.  The final objective of
these efforts is the specification of a series of near-term and long-term
options for reducing CH4 emissions from animal management systems.  These
options could then be examined in conjunction with options for reducing
emissions from other sources to identify a cost-effective approach for
reducing CH4 emissions globally.   Each of  the two areas of investigation is
discussed in turn.
3.1 IMPROVED EMISSIONS CHARACTERIZATION

    Emissions rates from various animals and animal management practices
must be better characterized.  Improved emissions estimates are needed in
order to:  assess anticipated changes in emissions over time; identify the
areas where emissions reductions efforts would be most promising; and
evaluate the emissions reductions that can be achieved with various.
options.

    In order to improve estimates as efficiently as possible, investigations
should be targeted at the key areas of inadequate data.  Several areas that
have been identified are summarized in Exhibit 3-1 and are described below.

    CHARACTERIZE THE ANIMAL POPULATIONS

    The population of managed ruminants throughout the world is reported by
the Food and Agriculture Organization (FAO) of the United Nations based on
reports to FAO from individual countries.  These data are important for
estimating CH4 emissions.   The accuracy of these data,  in-particular in
developing countries in Africa and Asia, remain to be examined.

    Additionally, to assess adequately recent and future trends in emissions
from these animals, differences and changes in animal characteristics
throughout the world must be estimated.  For example, the sizes of mature
cattle differ by a factor of at least 2 to 3 between developed and
developing countries.14  These  size  differences  are  indicative  of
differences in the amount of feed consumed by the animals.  An assessment
of animal sizes, amount of feed consumed, and feed characteristics would
assist in improving estimates of global CH4 emissions from animals.
     14   Of note is that cattle breeds found in developing countries  are  not
necessarily inherently smaller; instead, cattle growth is often stunted due
to inadequate nutrition.

-------
                                      32

                                 EXHIBIT 3-1

                   STEPS FOR IMPROVING EMISSIONS ESTIMATES
 1.
     AREA OF
  INVESTIGATION

Characterize animal
populations along
characteristics that
affect CH4 emissions.
2.
3.
Develop and evaluate
measurement
techniques.
Perform CHA emissions
measurements.
4.  Model Development.
                                         COMMENTS
 Good  animal population data are  required.   Defining
 categories of animals with similar characteristics
 will  facilitate subsequent analysis.  The numbers
 of animals are well described for most developed
 countries.  However, the majority of ruminants  are
 in developing countries where animal numbers are
 least well quantified.  Additionally, data  on
 animal characteristics (including animal waste
 management systems) are also required.  These data
 have  not been collected in a comprehensive  form for
 the major populations of animals in the world.

 Existing methods of measuring CH4 emissions
 directly from animals and from related sources  need
 to be evaluated.  New techniques are required to
 measure emissions rates from grazing animals and
 animals in developing countries.

 Adequate measurements of emissions rates from
 animals in developing countries appear to be
 lacking.  Additionally, measurements from some
 categories of animals in developed countries and
 from animal wastes are lacking.  The differences in
 CH4  emissions  among individual  animals,  and the
 factors that contribute to these differences (such
 as differences in the rumen microbial environment)
 remain to be measured and understood.

The animal population characteristics and the
 emissions measurements may be used for purposes of
developing a model of regional  and global
emissions.   The model may also  be a useful tool for
identifying the implications of alternative
approaches for reducing emissions.

-------
                                     33
    Similarly, animal waste management practices vary throughout the world.
The extent to which wastes are burned, spread as fertilizer, placed in
digesters, or disposed of in ponds must be evaluated.  A summary of animal
.waste management practices is required.

    One effective approach to developing this characterization is to define
a set of  "animal/management categories" that represent the range of animal
situations that exist throughout the world.  Such categories might include:
intensive dairy farming  in the U.S., using Holsteins and concentrate
feeding;  feedlot beef cattle in the U.S. on high grain diets; stocker
cattle in the U.S. in pastures and on ranges; cow/calf operations in the
U.S.; range/pasture cattle in South America; draft bovines in developing
countries that are fed rice and wheat byproducts; cattle in developing
countries that are managed by pastoralists and graze the available forage;
etc.

    The categories would differ along characteristics that influence CHA
emissions.  The key characteristics of each group could include:  number of
animals;  typical type of animal ownership; type and quantity of feed
consumed  (including key  feed characteristics); animal health services
typically available; products produced from the animals.

    DEVELOP AND EVALUATE MEASUREMENT TECHNIQUES

    Indirect calorimetry is the laboratory technique currently used to.
perform in-depth evaluations of the performance of alternative feeding
practices.  This technique involves placing an animal in confinement for a
period of several days,  and measuring the amount of inputs  (feed, oxygen,
carbon dioxide) and outputs (excretion, oxygen, carbon dioxide, CH4) from
the confinement chamber. Because CH4 is produced during digestion, CH4 is.
measured  as part of this technique.

    Because indirect calorimetry is primarily used to evaluate animal
energetics, its applicability for quantifying CH4 emissions must be examined
closely.  Several areas  that need to be addressed include:

    0     How representative are the CH4 emissions estimates for an
          individual animal of what might be expected for that individual
          animal in the field?  What factors could lead to the results not
          being representative, and are these factors important from the CH4
          point of view?  Examples of experimental factors that should be
          examined include:  (1) restrictions to the animal's intake to
          ensure that the experiment can be reproduced; (2)  stresses on the
          animal from being in confinement; (3) lack of environmental
          stresses on the animal  (e.g., lack of heat stress); (4)  lack of
          exercise (this  may be particularly important for evaluations of
          draft animals); and  (5) experiment duration.

-------
                                     34

    o    Have measurements been performed on an adequate set of animals and
         conditions?  A wide range of animal types and management practices
         exist.  Have sufficient measurements been performed across this
         entire range to provide a basis for quantifying emissions on a
         global basis?

    Because indirect calorimetry experiments are time consuming and costly,
other measurement techniques should be explored.  In particular, techniques
for measuring CH4 emissions from animals in developing countries,  and
grazing animals throughout the world need to be developed and evaluated.
For each of the major categories of animals (discussed above), an
appropriate measurement technique is required.

    Atmospheric sampling to estimate fluxes of CH* may be appropriate for
evaluating emissions from an entire animal management area, such as a
feedlot or a dairy.  The total emissions estimated with this technique
should be divided into those portions associated with rumen fermentation,
animal waste disposal, and other sources as appropriate.  Indirect
calorimetry or other methods may be appropriate for estimating the portion
associated with rumen fermentation.  Alternatively, the animals could be
removed from the management area and the measurements repeated.  To verify
these measurements, it would be useful to have techniques for evaluating
emissions from waste piles, waste lagoons, and soils.15

    Methods for evaluating emissions from grazing animals and small groups
of animals (e.g., one or two cows in a rural Indian household) remain to be
developed.  One .option that may prove useful is to collect air samples near
individual animals over an extended period of time.  In order to estimate
rates of CH4 emissions from the animal,  a method is required to adjust these
air samples for meteorological conditions.  The details of such an approach
remain to be developed.16
     15  As  described  in  the previous section, CH4 emissions have been
measured from waste lagoons.   The appropriateness of these methods for
characterizing the range of animal waste management practices remains to be
defined.

     16  One promising method for measuring CH4 emissions rates from
individual animals was suggested at the workshop.  The approach, involving
the use of tracers to estimate flux rates from animals, would allow CH4 and
carbon dioxide emissions rates to be measured inexpensively for individual
animals without the need to isolate the animals (i.e., without the use of a
calorimetry chamber.)  Such an approach, if it proves viable, will allow
numerous CH4 emissions estimates to be  developed for types of animals for
which estimates are not currently available (e.g., grazing animals and
animals in developing countries).   The details of the approach are
currently under development,  and will be presented in subsequent reports.

-------
                                     35

    As a complement to various techniques for measuring CH4  emissions
directly from animals, techniques could be explored to correlate emissions
from the animal to gases produced by samples of rumen contents, and to the
population (and species) of methanogens found in the samples.  By
developing such correlations, estimates of CH4 emissions could be produced
using laboratory analyses of rumen contents.  Such an approach, if
feasible, would greatly enhance the ability to estimate CH4  emissions  from
ruminants.

    PERFORM CH* EMISSIONS MEASUREMENTS

    Numerous measurements of CH< emissions from cattle and sheep in
developed countries have been made using direct and indirect calorimetry
techniques.  The representativeness of these measurements for the various
types of animal management practices must be assessed, and gaps in
measurements identified.  For example, indirect calorimetry experiments
have been performed on lactating dairy cows and finishing steers on various
types and levels of diets.  These data need to be consolidated and
evaluated.  Additionally, emissions estimates from brood cows, pasture
animals, and range animals are also required.

    The priorities for performing these measurements on various categories
of the animal population should be based on the importance of the category
in the overall population of animals and the availability and quality of
existing emissions data.  One obvious area where additional measurements
are required is CH4 emissions from animals in developing countries.  There
is uncertainty regarding their rates of CH4 production, in particular for
the types of diets they generally consume.

    Additionally, measurements are required to improve our -understanding of
the variations in emissions among individual animals that (except for their
apparent CH4 emissions rates) are similar, and the factors that influence
these variations.  The potential existence of host-animal traits that
influence methanogenesis by microbes in the rumen remains to be
investigated.  Potential mechanisms via which such traits could influence
rates of methanogenesis  (e.g., by influencing the populations of the
microbes) also are unknown at this time.

    Emissions measurements from animal wastes are also clearly required.

    MODEL DEVELOPMENT:  EMISSIONS AND ANIMAL MANAGEMENT

    By performing these various investigations, the basis of information
describing sources of CH4 emissions from ruminants will be improved
considerably.  These data could then be used to develop models for
quantifying emissions in various parts of the world and globally.  For
example, assuming that each of the major categories of animals will be
characterized in terms of key factors affecting emissions, these categories
can be used as the basis for an emissions analysis model.  The extent to

-------
                                      36

which  each  of  the  categories  is  found throughout  the world, and  the  likely
rates  of  emissions from  each  can be used  to estimate emissions.

     For example, Baldwin, Thornley, and Beever  (1987) present the results of
a dynamic simulation model of rumen digestive functions.  This model can be
used to evaluate the anticipated level of CH4 emissions for a wide, range of
diets.  Using  the  model  with  data describing the  populations of  animals
throughout  the world and the  diets they consume,  estimates of global CH4
emissions from animals could  potentially  be improved.  Of note is that the
suitability of the model for  analyzing animal diets in developing countries
remains to  be  assessed.

     In addition to improving  estimates of global  CH4 emissions from animals,
the  rumen digestion model could  be integrated into, or used in conjunction
with an economic model of the livestock sector.   Such a combined tool would
be useful for  evaluating the  implications of undertaking actions (such as
improving the  nutrition  of animals in developing  countries) to reduce CH*
emissions.  For example, the  alternative  diets would be evaluated with the
model  of  rumen digestion to estimate  CH< emissions and animal productivity
(e.g., amount  of milk produced over time).  The economic portion of the
model would evaluate the costs of changing the diets and the implications
of changes  in  animal productivity (e.g.,  increases in milk production) on
the  profitability  of the livestock sector, prices, and animal numbers.

     A possible framework for  such a model is presented in Exhibit 3-2.  As
shown  in  the exhibit, policies for reducing CH4 emissions would be specified
(e.g., subsidizing the use of certain feeds or feed supplements).  The
implications of these policies for animal management decision at the farm
level would be assessed  using estimates of prevailing input prices (e.g.,
prices for  feed) and output prices (e.g., prices  for milk and meat), as
well as estimates  of animal productivity  (i.e., the amount of inputs
required  to produce the  outputs).

     Based on the simulation of representative production decisions at the
farm level, aggregate levels  of production would be simulated, including
the  anticipated size of  the animal populations.  The extent to which these
aggregate levels indicated changes in the anticipated input and output
prices would also have to be  evaluated (e.g.,  if milk production increased
substantially, the price of milk could decline, depending on government
pricing policies).   If prices were anticipated to change, then the
simulated farm level decision would have to be revised based on the new
level of  input and output prices.  Such adjustments would be required until
convergence upon a solution was achieved.

    Such a model could be used to examine the  implications of various
policies under a wide variety of assumptions.   For example,  if feed
supplements were to be provided in developing countries to reduce CH4
emissions and increase animal productivity,  the model could be used to
assess  the conditions under which animal populations would likely increase

-------
                                     37

                               EXHIBIT 3-2

                        POSSIBLE FRAMEWORK  FOR A
                  LIVESTOCK-METHANE ASSESSMENT MODEL
Input Prices
     Feedback

 Feedback
                   Policy Specifications
                    Subsidies
                    Pricing
                    Trade
                    Other
                 Technology Assumptions
                         Output Prices
                Farm Level Decision Model
                 Farm Type 1
                 Farm Type 2
                 etc.
Aggregate Results
 National
 Regional
 Global
Feedback

	*
Major Outputs:
 Animal Population
 Animal Products

 Methane Emissions
 Costs of Policies

-------
                                     38

(thereby negating the emissions reduction impacts of the program) or
decrease (thereby reinforcing the emissions reduction goal).   Preferred
sets of policies that reduce emissions as well as improve the production of
animal products could potentially be identified.


3.2 OPTION IDENTIFICATION AND EVALUATION

    Options for reducing emissions must be identified and evaluated.  Given
the obvious importance of managed ruminants throughout the world, and their
unique ability to convert low quality inputs (e.g.,  agriculture byproducts
that cannot be digested by humans) into useful products, potential
technical and management options for reducing CH4 emissions must be
evaluated within the context of the overall agricultural and economic
systems in which they would be implemented.

    Potential options for reducing emissions should be evaluated in terms
of:

    o    Time frame:  the period when the option may become viable,  such as
         near term versus long term.

    o    Applicability:  the categories of animals for which the option may
         be used to reduce emissions, such as intensively managed dairy
         cows on concentrate feeds.

    o    Emissions Reduction:  the extent to which the option would reduce
         emissions.

    o    Impacts on Animal Productivity:  the manner in which implementing
         the option would affect the production of animal products.

    o    Costs:  capital costs, operating costs,  and other relevant costs.

    o    Implementation:  methods of implementing the option, including any
         special challenges posed, such as social constraints.

Based on an initial assessment of possible options,  the most promising
alternatives should be examined and their effectiveness should be
demonstrated.  Exhibit 3-3 displays a partial and preliminary assessment of
several options identified to date.  These options represent a rich set of
possible ways to reduce CH4 emissions,  increase animal productivity, and,  in
some cases, produce energy.  As such, some of the opportunities for
reducing CH4 emissions might in fact be economically viable  in their own
right, or might have very small overall costs.

    As shown in the exhibit, the options range from modifying the meat
grading system in the U.S. to changing feeding strategies in developing

-------
                                     39

                                 EXHIBIT  3-3

                 PARTIAL AND PRELIMINARY IDENTIFICATION AND
                EVALUATION OF CH4 EMISSIONS REDUCTION OPTIONS
   EMISSIONS REDUCTION
         OPTION
                      COMMENTS
Near Term Options

1.  Adopt alternative
    feeding practices to
    reduce methane
    emissions from
    animals in developed
    countries.
The use of alternative feeding practices is
applicable primarily to confined animals on
controlled rations, such as dairy cows and feed-lot
animals.  (The diets of grazing animals probably
cannot be modified easily in the short term.)
Given that methanogenesis is influenced by feed
characteristics, opportunities for reducing methane
emissions exist by modifying feeding practices.
For example, based on preliminary data it has. been
suggested that CH4 emissions associated with a
range of commonly used rations in the U.S. may vary
by as much as 50 percent.  If the "low CH4" rations
were emphasized in place of the "high CHA" rations,
CH4 emissions could be significantly reduced.   Such
an emphasis would, in all likelihood, increase
feeding costs.

This approach may be attractive in the near term
because it does not require a major departure from
the feeding systems currently in use.  The
implications of the modified feeding regimes for
animal productivity and feed costs remain to be
examined, however.  It is anticipated that
appropriate "low CH
-------
                                     40

                                 EXHIBIT 3-3

                  PARTIAL AND  PRELIMINARY IDENTIFICATION AND
                EVALUATION OF CH4 EMISSIONS REDUCTION OPTIONS
                                 (continued)
    EMISSIONS REDUCTION
          OPTION

Near Term Options  (continued)
                                              COMMENTS
 2.
 Increase animal
 productivity with
 hormones.
3.
 Increase  animal
 productivity with
 intact males.
4.
Modify the meat
grading system in the
U.S.
5.
Modify feeding
strategies in
developing countries
by using supplements
to correct nutrient
deficiencies.
 Hormone  implants  are currently available  for  non-
 lac tating beef animals  (these implants were
 recently banned in  the  EEC).  Emissions reduction
 is achieved  through faster weight gain and
 increased feed efficiency.  Bovine growth hormone
 (currently under  development) may increase milk
 production in lactating animals significantly,
 thereby  reducing  the size of the animal population.
 Food safety  must, of course, always be considered.

 The use  of intact males (foregoing castration) has
 been identified as  giving similar performance to
 the use  of hormone  implants.  The use of  intact
 males would  reduce  CHA emissions by promoting
 faster weight gain  and  increased feed efficiency.
 Issues associated with using intact males include
 managing more aggressive animals, changes in meat
 quality, and changes in taste.

 The current  meat  grading system in the U.S.
 attaches a premium  to finishing practices that
 produce  a given amount of fat.  It has been
 suggested that finishing practices that produce
 less fat may also produce less methane per amount
 of meat produced by reducing slaughter weights and
 finishing time.   Consequently, modifying the
 current grading system so that leaner meats were
 not given relatively lower grades could produce
 reductions in methane emissions.   This alternative
 should be examined within the context of the
 overall issues associated with meat grading.

This alternative is applicable to animals in
developing nations with nutrient deficiencies (such
as inadequate nitrogen).  Emissions reductions
associated with increased fermentation efficiency
are anticipated at the individual animal level.
For a population of animals, significant emissions

-------
                                     41

                                EXHIBIT 3-3

                 PARTIAL AND PRELIMINARY IDENTIFICATION AND
               EVALUATION  OF CH* EMISSIONS REDUCTION OPTIONS
                                (continued)
   EMISSIONS REDUCTION
         OPTION
                                             COMMENTS
Near Term Options (continued)
    Modify feeding
    strategies in
    developing countries
    (continued)
6.
Reduce CH4 emissions
from animal wastes.
reductions may be achieved through increases in
animal productivity (e.g.,  milk yield) and
reproductive efficiency (e.g., reduced inter-
calving interval).   Level of emissions reductions
remains to be quantified and demonstrated.
Molasses/urea blocks may be appropriate
supplements, or chicken litter.  Domestic
supplement manufacturing capability should .be
promoted.

This alternative is applicable to wastes currently
disposed of in lagoons or other anaerobic
environments.  Capturing CH4 emissions (e.g., for
energy) would significantly reduce emissions from
this source.  Efforts are currently under way to
develop techniques for harvesting CH* from
anaerobic waste lagoons.  These techniques involve
covering the lagoons with plastic covers and using
the CH4 to produce  electricity for use on site or.
for sale to the electric power grid.  Digesters to
produce and contain CH4 are also under development
and in use in the U.S. and elsewhere.

Integrated animal waste recycling facilities are
currently under development that not only produce
and capture CH4 in digesters, but also use the
carbon dioxide emissions produced from burning the
CH4 to "fertilize"  plant growth in greenhouses
and/or algae growth in ponds.  The plant and algae
products can then be marketed or used as animal
feeds on site.  The electric power produced by the
system (some of which can be sold to the power
grid) replaces the need to burn other fossil fuels.

Direct burning of wastes (i.e., for energy)  is also
a current waste management practice, although local
air quality impacts must also'be considered.

-------
                                     42

                                 EXHIBIT  3-3

                 PARTIAL AND PRELIMINARY IDENTIFICATION AND
                EVALUATION OF CH4 EMISSIONS REDUCTION OPTIONS
                                 (continued)
   EMISSIONS REDUCTION
          OPTION
                      COMMENTS
Long Term Options

7.  Modify feed
    characteristics or
    rumen processes to
    increase feed
    digestibility or
    reduce
    methanogenesis*

8.  Develop CH4
    inhibitors as feed
    additives.
9.  Improve reproductive
    efficiency to reduce
    brood herd
    requirements.
Alternatives under consideration include
genetically engineered bacteria to pre-treat feed
or to be introduced into the rumen.  Modifications
in rumen digestion could lower emissions.  This
alternative is most likely to be applicable to
intensively managed animals in developed countries.
In the past, efforts have been made to identify and
develop CH* inhibitors.   Although several compounds
that effectively inhibit methanogenesis in vitro
have been identified, most have not been marketed
as a CH4 inhibitor due to insufficient improvement
in feed efficiency in the whole animal.  lonophore
feed additives (discussed above) have been
demonstrated to improve feed efficiency and reduce
CH4 production,  and are  widely used in beef animals
in the U.S.

The development of new options is uncertain, and
possibly unlikely.  The CH4 inhibitor would have to
provide an additional hydrogen sink assuming that
less CH4 is created.   Additionally,  the inhibitor
would have to be acceptable from the viewpoint of
human, animal, and environmental risks.

This alternative is applicable to areas with brood
herds.  In developing countries, nutritional
management programs to increase reproductive
efficiency (e.g., toward one cow per calf per year)
may be appropriate.  In all circumstances, it is
important to ensure sufficient nutrition to new
born calves to ensure survival.

-------
                                     43

                                 EXHIBIT  3-3

                 PARTIAL AND PRELIMINARY IDENTIFICATION AND
                EVALUATION OF CH4 EMISSIONS REDUCTION OPTIONS
                                 (continued)
   EMISSIONS REDUCTION
          OPTION

Long Term Options (continued)
                      COMMENTS
10. Breed animals that
    are low CH4
    producers.
The objective of this alternative is to identify
whether there are heritable genetic characteristics
that account for variability in CH<, emissions among
individual animals, and to breed animals that are
low CH4 producers by taking advantage of these
characteristics.  The existence of such traits
remains to be demonstrated, although preliminary
data indicate that such traits may exist.

In order for such an approach to work, the relevant
animal characteristics and the mechanisms via which
the characteristics influence methanogenesis by
microbes in the rumen would have to be identified.
If such characteristics could be identified, and if
they were not undesirable for other reasons (e.g.,
also being associated with low rates of digestion),
animals could be selected for the "low CH4"
characteristics.  If the characteristics were
undesirable for other reasons, the mechanisms via
which the characteristics influenced methanogenesis
by microbes could potentially-be exploited.

This approach remains speculative at this time
because host-animal characteristics and mechanisms
that influence methanogenesis by.microbes in the
rumen have not yet been identified.  One avenue for
initiating such research may be to take advantage
of the considerable amount of existing data on the
breeding histories of dairy cows and some beef
animals in the U.S.  These data may enable
hypotheses about heritability to be tested by
measuring CH4 emissions from a large number of cows
and examining whether lineage explains any of the
variance that exists among the individuals.  In
order to be successful, reliable methods for
inexpensively measuring CH4 emissions from large
numbers of animals are required.

-------
                                     44

.countries.   These,  and additional  options must be described more  completely
 and evaluated to  identify  the most cost  effective measures for  reducing
 emissions.

     Of particular importance is identifying and involving the institutions
 and organizations that could play  a role in evaluating and implementing the
 various  options.  Such organization would include government agencies
 (e.g., the Agriculture Research Service  of the U.S. Department  of
 Agriculture),  universities, and private  companies (e.g., those  producing
 ionophores and other feed  additives).  Major organizations that influence
 agriculture  practices  in developing countries include:

     o    United Nations (e.g., through the Food and Agriculture
         Organization  and  other programs);

     o    World Bank and other development banks (e.g., as a funding
         source);

     o    Consultative  Group on International Agricultural Research (e.g.,
         through  the International  Livestock Center for Africa  (ILCA),  and
         the  International Laboratory for Research on Animal Diseases
         (ILRAD))

     o    Vinrock  International, a private philanthropic non-profit
         organization  providing technical assistance worldwide;

     o    Agricultural  Cooperative Development International,  focused on
         assisting and developing agricultural cooperatives worldwide;

     o    Volunteer International Technical Assistance, focused  on providing
         technical agricultural assistance worldwide;  and

     o    Programs of various national governments,  such as the U.S. Agency
         for International Development (AID).

     Coordination among  these and other key institutions is required to
ensure the timely development of options for reducing CHA  emissions.   The
creation of an ad hoc working group for communication of ongoing and
planned activities and  tracking progress is planned.

-------
              4.  EPA OFFICE OF AIR AND RADIATION PROGRAM PLAN
    This chapter presents a plan for undertaking analyses to:  (1) improve
current estimates of CH4 emissions from animals,  and in particular
ruminants; and (2) identify, evaluate, and implement options for reducing
emissions from this source.  The outline for this plan is based on the
findings presented at the beginning of this document.

    The following program elements are planned:

    o    Ad Hoc Group.  An Ad Hoc group would track and coordinate' the
         analyses and investigations needed to better quantify CH4 emissions
         from animals and options for reducing the emissions.  The Ad Hoc
         group should include researchers and practitioners in the fields
         of animal science, animal management, atmospheric science, and
         others as appropriate; different people would participate at
         different times.  The mission of the Ad Hoc group should include:

                   develop a plan for various nations and international
                   institutions for solving the problem of CH4 emissions
                   from animals;

                   provide peer review on analyses and investigations
                   performed;

                   identify parties who may be appropriate funding
                   sources for these  investigations;

                   identify organizations who may be able to perform
                   investigations.

    o     Emissions Characterization.  A series of investigations will be
          identified that will improve our understanding of the rates of CH4
          emissions from animals.  These investigations will  include the
          following.

                   Identify those animals for which  emissions estimates
                   are  particularly uncertain.   For  these groups  of
                   animals, develop and employ techniques for measuring
                   their emissions within the next year.  It is likely
                   that animals  in developing countries will fall into
                   this category.

                --  Develop and implement techniques  for measuring CH4
                   emissions  from animals under  field  conditions.
                   Opportunities include atmospheric measurements near
                   animal  management  facilities  as well  as  techniques
                   for  measuring emissions  from  individual  animals.

-------
           Characterize  the  current  and potential  future
           populations of animals along dimensions that
           influence  CH4 emissions and opportunities for
           reducing emissions, such  as:  quantity  and  quality of
           feed consumed; waste management practices used;
           trends  in  population size and characteristics;
           ownership  characteristics; and products produced.

           Develop data  on the current and anticipated methods
           of managing animal wastes in the U.S. and around  the
           world.  Evaluate  existing measurements  of CH4
           emissions  from animal wastes, and identify  gaps in
           measurements.  Develop and employ measurement
           techniques to fill these  gaps.  It  is likely that
           most waste management systems should be measured.

           In conjunction with field measurement efforts, develop
           and employ modeling approaches for  assessing CH4
           emissions  from animals and animal wastes.   In
           particular, existing data on CH4 yields and kinetics  in
           wastes  should be  examined to improve estimates of
           potential  emissions.

Option  Identification and Evaluation.  Both near-term and long-
term  options for  reducing CH4 emissions must be examined.  These
.assessments  must  include evaluations of:  emissions reduction
potential; costs; social and other  barriers or constraints;  and
impacts on the quality  and  safety of the food supply.  Activities
will  include the  following.

           Develop and employ modeling techniques  to evaluate fully
           alternative approaches for reducing methane emissions.
           Analysis of the implications for animal populations will
           be included in this analysis, as well as assessments  of
           emissions  from individual animals.

           Evaluate the  potential advisability, effectiveness,
           and cost of modifying feed practices and/or the meat
           grading system in the U.S. to reduce CH* emissions.

           Evaluate the  potential advisability, effectiveness,
           and cost of using nutrient supplements  as a means  of
           reducing CH4 emissions among populations of animals
           in developing countries.

           Evaluate the  implications of emerging technologies
           for CH4 emissions, such as bovine growth hormone.

-------
                                     47

                   Identify and evaluate options for reducing or
                   recovering CH4 emissions from animal wastes.

                   Identify promising long term approaches for reducing
                   CH4 emissions and initiate investigations into these
                   approaches.  Examples may include:  developing
                   improved CH4 inhibitors; identifying animal traits
                   that influence methanogenesis by microbes in the
                   rumen; modifying feed characteristics to increase
                   digestibility or reduce methanogenesis.

    These major program elements should be implemented within the coming
year.  For each, a more detailed plan will be developed that includes:
detailed descriptions of the proposed investigations; schedule for
undertaking the investigations, and approximate level of resources that is
appropriate; identification of individuals and institutions that could
perform the investigations; identification of appropriate funding sources
(e.g.,  EPA, Department of Agriculture (through CSRS or ARS), NASA, NSF,
and others); and methods for obtaining funding.

    As soon as practicable, validations and demonstrations of the various
CH4 emissions reduction alternatives will be organized.  Finally,
alternatives for implementing these emissions reduction methods will be
developed through national and international institutions and international
forums.

-------
                                                   >•';.-£ !;9v:a!-c. Belong To:

                                  REFERENCES        ^ '/, V^-,^; S-vV i TS-793)
                                                               c 20460
 American Meat Institute,  Meat  Facts.  Washington,  D.C.,  1987.

 Baldwin, R.L.  and M.J.  Allison,  "Rumen Metabolism,"  Journal  of Animal
 Science. Vol.  57,  1983, pp.  461-477.

 Baldwin, R.L. ,  J.H.M. Thornley,  and  D.E.  Beever,  "Metabolism of the
 Lactating Cow.   II.  Digestive  Elements of a Mechanistic Model," Journal of
 Dairy Research.  Vol.  54,  1987, pp. 107-131.

 Blake,  D.R.  and F.S.  Rowland,  "Continuing Worldwide  Increase in
 Tropospheric Methane, 1978  to  1987,"  Science. March  4,  1988.

 Blaxter, K.L.  and J.L.  Clapperton, "Prediction  of the Amount of Methane
 Produced by  Ruminants," British  Journal of Nutrition. Vol. 19,  1965,  pp.
 511-522.
                                                                *
 Bolle,  H.-J., W.  Seller and  B. Bolin,  "Other Greenhouse Gases and Aerosols:
 Assessing Their  Role  for  Atmospheric  Radiative  Transfer,"  in:   Bolin,  B. ,
 B.R.  Doos, B. Warrick and D. Jager (eds.), The  Greenhouse  Effect Climatic
 Change  and Ecosystems. John  Wiley &  Sons:  New  York, 1986.

 Crutzen,  P.J.,  I.  Aselmann,  and  W. Seiler, "Methane  Production by Domestic
 Animals,  Wild Ruminants,  Other Herbivorous Fauna,  and Humans,"  Tellus .  38B,
 1986, pp.  271-284.

 Crutzen,  Paul J.,  "Role of the Tropics  in Atmospheric Chemistry," in:
 Dickinson, R.  (ed.),  Geophysiology of Amazonia. John Wiley & Sons:  New
 York, 1987.

 Dunkley,  W.L. ,  "Reducing  Fat in  Milk  and Dairy  Products -  Introduction,"
 Journal  of Dairy  Science. Vol. 65, 1982, p. 442.

 Ensminger, M.E., The  Stockman's  Handbook. The Interstate Printers &
 Publishers,  Inc.:  Danville, Illinois,  1983.

 Ensminger, M.E., Beef Cattle Science. The Interstate Printers & Publishers,
 Inc.:  Danville,  Illinois, 1987.

 EPA,  Policy  Options for Stabilizing Global Climate. Draft Report to
 Congress,  EPA:  Washington, D.C., 1989.

 EPA, Assessing the Risks of Trace Gases That Can Modify the Stratosphere.
 EPA:  Washington, D.C. ,  1987.

 Fallert, R. , T. McGuckin,  C. Betts,  and G. Bruner  (1987),  bST and the Dairy
 Industry: A National. Regional and Farm-Level Analysis. U.S.  Department of
Agriculture, Washington, D.C. ,  p. 114.

-------
                                     49
FAO, FAQ Production Yearbook. Food and Agriculture Organization of the
United Nations, Rome, selected years.

FAO, Livestock Numbers Database Printout, Food and Agriculture Organization
of the United Nations, Rome, Italy, 1988.

Goreau, T. and W.Z. Mello, "Effects of Deforestation on Sources of
Atmospheric Carbon Dioxide, Nitrous Oxide, and Methane from Central
Amazonian Soils and Biota During the Dry Season: A Preliminary Study,"
Proceeding of the Workshop on Biogeochemistry of Tropical Rain Forests:
Problems for Research. Sao Paulo, Brazil, World Wildlife Fund-U.S., Centre
de Energia Nuclear na Agricultura (CENA), and Universidade de Sao Paulo,
1985, pp. 51-66.

Hansen, J., et al.. "The Greenhouse Effect:  Projections of Global Climate
Change," in:  Titus, J.G. (ed.), Effects of Changes in Stratospheric Ozone
and Global Climate. U.S. EPA and UNEP:  Washington, D.C., August 1986, p.
206.

Hill, D.T., "Methane Productivity of the Major Animal Types," Transactions
of ASAE. 1984, p.  530.

Hoffman, J.H., and M.J. Gibbs, Future Concentrations of Stratospheric
Chlorine and Bromine. U.S. Environmental Protection Agency:  Washington,
D.C., EPA 400/1-88/006, August 1988.

Khalil, M.A.K., and R.A. Rasmussen, "Trends of Atmospheric Methane:  Past,
Present, and Future," in:  Proceedings of the Symposium on C02 and Other
Greenhouse Gases.  Brussels, Belgium, November 1986.

Mix, L.S., "Potential Impact of the Growth Hormone and Other Technology on
the United States  Dairy Industry by the Year 2000," Journal of Dairy.
Science. Vol. 70,  1987, pp. 487-497.

Moe, P.W. and H.F. Tyrrell,  "Methane Production in Dairy Cows," Journal of
Dairy Science. Vol. 62, 1979. pp. 1583-1586.

NMPF, Dairy Producer Highlights. National Milk Producer Federation,
Arlington, Virginia, 1987.

NRC, Nutrient Requirements of Beef Cattle. National Academy Press:
Washington, D.C.,  1984.

NRC, Nutrient Requirements of Dairy Cattle. National Academy Press:
Washington, D.C.,  1988.

Preston, T.R. and  R.A. Leng, Matching Ruminant Production Systems with
Available Resources  in the Tropics and Sub-tropics. Penambul Books,
Armidale, New South Wales, Australia, 1987.

-------
                                     50
Rojko, A.S., et al.,  Alternative Futures for World Food in 1985. Foreign
Agricultural Economic Report No. 146, GPO:  Washington, D.C.,  April 1978.

Rumpler, W.V., D.E. Johnson, and D.B. Bates, "The Effects of High Dietary
Cation Concentration on Methanogenesis by Steers Fed Diets With and Without
lonophores," Journal of Animal Science. Vol. 62, 1986, pp. 1737-1741.

Safley, M.L., and P.W. Westerman, "Biogas Production from Anaerobic
Lagoons," Biological Wastes. Vol. 23, 1988, pp. 181-193.

Seiler, W.,  "Contribution of Biological Processes to the Global Budget of
CH4 in the Atmosphere,"  in:  Klug,  M.  and C.  Reddy (eds.),  Current
Perspectives in Microbial Ecology. 1984.

WHO (World Meteorological Organization), Atmospheric Ozone 1985. NASA:
Washington, D.C.,  1986.

-------
                                 APPENDIX A

                   GLOBAL METHANE  EMISSIONS FROM RUMINANTS
    This appendix summarizes several of the published estimates of methane
(CH4) emissions from ruminants.   First,  the primary factors affecting
methanogenesis in ruminants are discussed.  Then, several of the major
estimates of emissions are summarized.
A.I FACTORS AFFECTING METHANOGENESIS

    The rumen is the unique physiological characteristic of ruminants that
allows them to digest the energy in forages that cannot be consumed by
humans.  These animals (mostly cows, buffalo, sheep, goats, and camels, but
also including deer, reindeer, caribou and others) are characterized by a
large "fore-stomach" or rumen.  For example, a cow's rumen may have a
volume of 150 liters, whereas a pig (a non-ruminant monogastric animal)1 has
a stomach with a size of about 6 to 8 liters.2  This large "stomach" allows
large quantities of forages to be consumed, and provides a location for the
forages to be fermented.  Other non-ruminant herbivores (e.g., horses,
rabbits, guinea pigs, hamsters) are able to consume forages in amounts
between ruminants and non-ruminant monogastries due to an enlarged cecum
and large intestine (Ensminger (1983). p. 21).

    Within the rumen, microorganisms play an important role in digestion.
Over 200 species and strains of organisms have been identified to date,
although a smaller number dominate (Baldwin and Allison (1983), p. 462).
These organisms form a complex ecology that includes both competition and
cooperation.  The organisms are important for fermenting the primary feed
constituents (carbohydrates, lipids, urea, and proteins) into secondary
products that can be catabolized further and subsequently used by the
animal.  The population mix of the bacteria is influenced by the diet
consumed by the animal.

    Rumen methanogenic bacteria are the source of CH4 produced in ruminants.
Although these bacteria are a very small fraction of the total population
of microorganisms in the rumen, they play an important role in the complex.
rumen ecology.  While methanogens can convert acetate (a fermentation
product produced in the rumen) to CH4 and C02,  this  pathway for CH*
production in the rumen is believed to be of minor importance (Baldwin and
     1  Other non-ruminant monogastric animals include inter alia:  dogs, cats,
monkeys, and humans.

     2  A steer's rumen is about 20 times larger than a pig's stomach, even
though a steer weighs only about 5 times more than a full-grown pig (Ensminger
(1983), p. 216).

-------
                                     A-2

Allison  (1983), p.  469).   Instead,  the  conversion of hydrogen  (H2) or
formate  and  C02 (produced by other  fermentative bacteria) is believed to be
the primary  mechanism by which methanogenic bacteria produce CHA in
ruminants.

    As shown in Exhibit A-l, on a  "whole-animal" basis,  the manner in which
the energy intake of an animal is  utilized can be defined as follows:

    o    gross energy is the total  energy intake by the  animal, where the
         energy content of  the feed is  defined in terms  of its total
         combustible energy (i.e.,  the  energy it releases when it is
         burned);

    o    digestible energy  is the  gross energy intake minus the energy
         eliminated in feces;

    o    metabolizable energy is the digestible energy minus the energy
         eliminated in urine and gas (CH4);  and

    o    net  energy is the  metabolizable energy minus the increment of heat
         produced by the animal.

Net energy is a measure of  the extent to which feed contributes to the
maintenance and growth of an animal.3  The net energy received by an animal
from a diet will depend on  the energy intake level of the diet* and the form
of the energy (i.e., the mix of feeds that comprise the  diet).  Some feeds
(e.g., grains) are more "digestible" than others (e.g., hay), so that more
of the gross  energy is converted to digestible energy (and subsequently,
net energy).   Evaluations of alternative diets and diet  formulation
techniques continue to be an important and active area of research in
animal management.

    Of note is that the analysis of ruminant feeding practices in terms of
these energy  quantities may not be appropriate for many animals in
developing countries whose  diets are deficient in one or more important
     3  Maintenance energy is the amount of energy that will result in no loss
or gain in total body energy, and is defined as the amount of energy
equivalent to the fasting heat production (NRC (1984), p. 3).  For U.S. beef
cattle, maintenance energy requirements are estimated to be about 77
kilocalories times the animal weight in kilograms raised to the 0.75.power.
This requirement varies on the order of 3 to 14 percent for different breeds
and sex.  For U.S. dairy cattle, the maintenance requirement is on the order
of 12 percent higher than for beef cattle.

     *  The energy intake  level  of a diet is often described as the ratio of
the metabolizable energy intake to the maintenance energy requirement.  For
example, the level of the diet fed to lactating dairy cows in the U.S. may be
on the order of two to three times the maintenance requirement.

-------
                                 A-3


                             EXHIBIT A-l


                   ENERGY UTILIZATION  IN  RUMINANTS
     GROSS ENERGY
   Fecal Energy
                 DIGESTIBLE ENERGY
               (Roughly comparable to TON)
           Urinary and
          Combustible Gas
             Energy
METABOLIZABLE ENERGY
                                       NET ENERGY
                        Net Energy Maintenance
                        Net Energy Production
Source:     Ensminger, M.E., The  Stockman's Handbook. The  Interstate
            Printers & Publishers,  Inc.:   Danville, Illinois,  1983, p.

            245.

-------
                                     A-4

nutrients  (such as nitrogen).  In these cases, the efficiency of the
fermentation in the rumen may be reduced significantly, so that the animal
may derive much less useful energy from the feed than would be indicated by
the feed's physio-chemical properties.  Due to the nutrient deficiencies,
increasing feed intakes (without correcting the nutrient deficiencies)
would not necessarily increase the amount of useful energy "delivered" to
the animal.  Correcting the nutrient deficiencies would improve rumen
fermentation efficiency so that current levels of feeding would provide
more useful energy to the animal.  The use of this approach for animal
feeding in developed countries has been suggested as well, although its
applicability has not been demonstrated.

    The extent of methanogenesis in individual ruminants has been estimated
by various authors.5  Blaxter and Clapperton (1965)  reviewed the results of
615 closed-circuit respiration indirect calorimetry experiments on sheep
and cattle performed over a period of 10 years.  Based on an analysis of
the results for 48 different diets in 391 different experiments on 4-5
sheep for various levels of feeding, Blaxter and Clapperton identified feed
digestibility and level of intake to be important factors influencing the
extent of methanogenesis in the rumen, and developed the following equation
to describe CH4 production:

          Ym - 1.30 + 0.112 D + L (2.37 - 0.050 D)

where Ym is the CHA yield (megajoules (MJ)  of CH4 produced per  100 MJ  of
gross energy feed intake), L is the ratio of net energy intake to
maintenance energy requirements (e.g., two times maintenance), and D is the
percent digestibility of the feed (e.g., 50 percent).  The CH4 yield
estimated with this equation can be interpreted as the percent of gross
energy intake that is converted to CH4 within the  animal.  The digestibility
of the diets examined ranged from poor hay (54 percent digestible at
maintenance) to sugar-beet pulp (87.2 percent digestible at maintenance).
The levels of the diets ranged from one to three times maintenance.

    Exhibit A-2 presents estimates of CH4 yield using the  Blaxter and
Clapperton equation.  CH4  yield increases with increasing  digestibility for
feed levels near maintenance.  Conversely,  at feed levels greater than 2.4
times maintenance CHA yield decreases with  increasing digestibility.   Of
note is that the results for diets exceeding 2.5 times maintenance are
based on relatively fewer observations, and may be less certain.  Based on
a comparison of methanogenesis in cattle and sheep for seven diets,  Blaxter
and Clapperton state that these results (based on data for sheep) may also
be appropriate for evaluating methanogenesis in cattle.
        See for example:   Blaxter and Clapperton (1965)  and Moe and Tyrrell
(1979) and Rumpler, Johnson, and Bates (1986).

-------
                                    A-5
                                EXHIBIT A-2

                       METHANE YIELD AS A FUNCTION OF
               DIGESTIBILITY AND LEVEL OF GROSS ENERGY INTAKE
                                  (percent)
                               Level of  Intake

                     1.0         1.5         2.0         2.5      '   3.0

   Digestibility

         50

         60

         70

         80

         90
6.8
7.4
8.0
8.6
9.2
6.7
7.1
7.4
7.8
8.2
6.6
6.8
6.9
7.0
7.1
6.6
6.4
6.3
6.2
6.1
6.5
6.1
5.8
5.4
5.0
    Estimates based on an analysis of 391 experiments on sheep.  Methane
yield is the percent of feed energy that is converted to methane.  Level of
intake is measured as the ratio of total gross energy intake to total gross
energy intake needed to meet maintenance energy requirements (e.g.,  total
gross energy intake equals two times maintenance requirements).
Digestibility in percent.
    Source:  Blaxter, K.L. and J.L. Clapperton, "Prediction of the Amount of
Methane Produced by Ruminants," British Journal of Nutrition. Vol. 19,
1965, pp. 511-522.

-------
                                      A-6

     In reviewing the estimates in Exhibit A-2, care must be taken because  it
  is unlikely that animals eating near maintenance will be eating highly
  digestible feed.  Animals near maintenance  (e.g., draft animals in
  developing nations) would most likely be eating crop residues, such as rice
  stalks, which have low levels of digestibility (e.g., less than 50
  percent).  Similarly, it is unlikely that animals eating at high levels of
  intake would be eating low digestible feed.  The animal would be unable to
  consume enough low digestible feed in order to reach three times
  maintenance.

     Another model that predicts CH4 emissions was derived by Moe and Tyrrell
  (1979).  A total of 404 total energy balance trials based on indirect
  calorimetry were performed on Holstein dairy cows.  Unlike the Blaxter and
  Clapperton model, Moe and Tyrrell estimate CH4 emissions as a function of
  the amounts of the following feed constituents consumed:  soluble residue,
  hemicellulose, and cellulose.  Feeds vary in their content of these
  constituents.  CH4 production was found not to be dependent on the intake of
  other feed characteristics such as crude protein, ether extract, and
  lignin.  Based on their experiments Moe and Tyrrell estimated the following
"equation:

      CH«  (MJ/day)  - 3.406  + 0.510 soluble residue (kg fed)
                           + 1.736 hemicellulose (kg fed)
                           + 2.648 cellulose (kg fed).

 The analysis by Moe and Tyrrell is distinguished from the earlier work by
 Blaxter and Clapperton in that it relates methanogenesis to feed
• characteristics in addition to level of intake and digestibility.

     Most ruminants in the world are in developing countries where levels of
 intake and feed characteristics are very different from the diet of U.S.
 dairy cows.   In general,  ruminants in developing countries eat agriculture
 by-products with low digestibility (e.g., wheat straw,  rice straw,  and
 sugar beet tops),  with levels of intake near maintenance requirements.  For "
 these animals on relatively poor diets,  Preston and Leng (1987)  indicate
 that methane yields may be in the 9 to 12 percent range, with the level
 influenced by the adequacy of ammonia levels in the rumen (Preston and Leng
 (1987),  p.  40).   In their assessment,  based on stoichiometric
 considerations,  inadequate ammonia levels are related to inefficient rumen
 fermentation and increased CHA production.   If  this  is the  case,
 supplementing diets of these animals  with urea or poultry litter (good
 sources of rumen ammonia)  could increase rumen fermentation efficiency and
 reduce CH4 emissions.

     It is  clear  from these  various  assessments,  that CH4 emissions from the
 digestive  tracts  of ruminants  depends  on the amounts and types of feeds
 consumed by  the  animals.   Because energy requirements for maintenance  and
 growth depend on  additional factors  (e.g.,  breed,  sex, stage  of  lactation,
heat  stress),  the  extent of methanogenesis  may  also be influenced by

-------
                                    A-7

additional factors.  However, if such influences exist, they have not been
sufficiently significant to be Identified to date.

    Given these general factors affecting methanogenesis in ruminants, the
general approach to estimating CH4 emissions from this source has been to:

    o    identify the average amounts of feed consumed by various types of
         ruminants;

    o    adopt an estimate of the portion of feed energy consumed that is
         converted to CH4 (i.e.,  a methane yield);

    o    estimate a rate of annual CH4 emissions per animal by multiplying
         the methane yield by the annual feed consumption; and

    o    multiply the emissions rate per animal by the number of animals.

Estimates published by various authors are presented below.


A.2 GLOBAL EMISSIONS ESTIMATES

    Several authors have presented estimates of global CH4 emissions from
animals6.  Most estimates, however, can be traced back to only a few
sources:

    o    Ritzman and Benedict (1938):  Hutchinson (1948) and Blake  (1984)
         base their estimates on CH4 production rates from domestic animals
         on work presented by Ritzman and Benedict  (1938).   Ehhalt  (1974)
         and Koyama (1963) cite emissions estimates from Hutchinson.
         Hudson and Reed  (1979) cite an emissions estimate from Ehhalt.
         Khali1 and Rasmussen (1983) cite an emissions estimate from Hudson
         and Reed;

    o    Crutzen and Seller:  Crutzen and Seller  contributed to various
         estimates including:  Crutzen et al.  (1986);  estimates by  Seiler
         in Bolle  et al.  (1986); and Seiler  (1984).   Lerner et al. present
         estimates based  on CH4 production rates from Crutzen et al. (1986).
         Sheppard  et al.  (1982) cite emission  data  from a personal
         communication with Crutzen.

In addition, both  Ehhalt  (1974) and Blake  (1984)  cite Woodwell (1970)  as
the source of total plant material consumed by herbivores.
     6  See, for example:  Blake (1984); Bolle, Seiler and Bolin (1986);
Crutzen, Aselmann and  Seiler  (1986);  Ehhalt  (1974); Hudson and Reed  (1979);
Hutchinson  (1948); Khali1  and Rasmussen (1983); Koyama  (1963); Lerner,
Matthews and Fung (1988);  Seiler  (1984); Sheppard et al.  (1982).

-------
                                    A-8
    This section presents estimates, of global CH4  emissions  from ruminants
published in three of the above mentioned studies:

    o    Crutzen et al. (1986):  The study by Crutzen et al. is the most
         comprehensive description of CH4 emissions  to date.

    o    Blake (1984):  Blake used data from Woodwell (1970) and Ritzman
         and Benedict (1938) to estimate global emissions via two different
         methods.

    o    Lerner et al. (1988):  Lerner et al. gathered detailed data on
         animal populations and estimated CH4 emissions by latitude and
         longitude.

    Exhibit A-3 summarizes the estimates presented by these  authors.  Each
is discussed in turn.
    CRUTZEN ET AL.

    Crutzen et al. estimated total annual CH4  emissions by:   (1)  estimating
the average CH4 production per animal  in various  regions  of  the world;  (2)
multiplying the average emissions per animal by the total population of
animals in each region; and (3) summing across the regions.   The average
CH4 production rates were derived from the Blaxter and Clapperton (1965)
relationship (discussed above) and other sources.  The population data for
each region were taken primarily from FAO statistics.  This  publication by
Crutzen et al. is the most comprehensive description of  CH4  emissions from
ruminants to date.  A copy of this publication is included as Appendix C.

    Exhibit A-4 presents the Crutzen at al. estimates for CH4 emissions.
Bovines, including all cattle used for beef production,  milk production,
and draft animals, are the largest source, accounting for 70 percent of the
total estimate and about 75 percent of the CH4 emissions  from managed
ruminants.

    As shown in Exhibit A-4, Crutzen et al. divided the  world into developed
and developing countries for evaluating bovine emissions.  The emissions
estimates for each is based on the following:

    o    Developed Countries.  The emissions rate for all developed
         countries plus Brazil and Argentina is based on evaluations in
         each of the following countries:

               United States and Canada.  This region has about  114 million
               cattle,  which can be divided into  three main  groups, each
               with different CH4 emissions rates:   dairy cows; beef cattle
               on feed; and beef cattle on the range.  In the U.S., these

-------
                                     A-9
                                 EXHIBIT  A-3
                     SUMMARY OF GLOBAL METHANE  EMISSIONS
     AUTHOR
Crutzen et al.
EMISSIONS
 (Tg/yr)

   77.7
Blake
   71-160
Lerner et al.
  75.8
                    COMMENTS
Includes:  domestic ruminants (cattle, buffalo,
sheep, goats, camels); domestic pseudoruminants
(horses, pigs); wild ruminants and large
herbivores; and humans. Approximately 95%
associated with domestic ruminants, of which
about 75% is associated with cattle.  Reported
uncertainty of ±15% for domestic animal and
human total.  Average estimate of CH4
production from wild herbivores is 4 Tg,
ranging from 2 to 6 Tg.  Animal population data
from FAO.  Feed energy intake and methane yield
from bovines vary for the following regions:
U.S. dairy cows; U.S., Argentine and Brazilian
range cattle; other developed countries; and
developing countries.

Includes:  domestic ruminants (cattle, sheep,
goats); domestic pseudoruminants (horses); wild
ruminants and large herbivores; and herbivorous
insects (i.e., termites).  Lower estimate based
only on domestic ruminant measurements.  Upper
limit based on 10% of total dry plant matter
produced consumed by all herbivores, assuming
43% is eaten by cattle and converted to CH4
with a 2% efficiency, and the remaining 57% is
converted to CH4 with a 1% efficiency.  Animal
population data from FAO.

Includes:  domestic ruminants (cattle, buffalo.
sheep, goats, camels); domestic pseudoruminants
(horses, pigs); and wild ruminants (caribou).
Global animal population density on a 1°
latitude by 1° longitude grid from data
compiled from FAO and individual country
statistics.  Methane emissions by latitude and
longitude based on emissions rates per animal
from Crutzen et al. (1986).  Over 75% of
emissions are associated with cattle.  About
55% of total emissions is concentrated from
25°N to 55°N latitude.

-------

ESTIMATES OF
A- 10
EXHIBIT A-4


METHANE EMISSIONS FROM ANIMALS
REPORTED BY CRUTZEN
ANIMAL TYPE AND REGIONS
Cattle in developed countries,
Brazil and Argentina
Cattle in developing
countries :
Buffalos
Sheep in developed countries
Sheep in developing countries
and Australia
Goats
Camels
Pigs in developed countries
Pigs in developing countries
Horses
Mules, Asses
Humans
Wild ruminants and large
herbivores
TOTAL
POPULATION
(Million)
572.6
652.8
142.1
399.7
737.6
476.1
17.0
328.8
444.8
64.2
53.9
4669.7
100-500

ET AL.
CH4 PRODUCTION
PER INDIVIDUAL
(kg/year)
55
35
50
8
5
5
58
1.5
1.0
18
10
0.05
1-50


CH4 PRODUCTION BY
TOTAL POPULATION
(Tg/year) **
31.5
22.8
6.2
3.2
3.7
2.4
1.0
0.5
0.4
1.2
0,5
0.3
2-6
75.7-79.7
    ** Total estimate for emissions from domestic animals (cattle, •feuffalo, sheep,
goats, camels, pigs, horses, mules and asses) has an uncertainty of ±15 percent.

    Source:  Crutzen, P.J., I. Aselmann, and W.  Seiler, "Methane Production by Domestic
Animals, Wild Ruminants, Other Herbivorous Fauna, and Humans," Tellus.  38B, 1986, pp.
271-284.

-------
                                    A-ll

               groups  represent  about  10 percent,  12.5 percent and  77.5
               percent of the  total  respectively.7  Canada is assumed to
               have  approximately  the  same  distribution.

               Crutzen et al.  estimate the  gross energy  intake and  CH*
               production for  dairy  cows in the U.S.  and Canada  to  be about
               230 MJ  per day  and  84 kg per year (based  on a methane yield
               of 5.5  percent).  For cattle on feed and  on range the
               estimates are 150 and 110 MJ/day of gross energy  intake,
               respectively, and 65  kg and  54 kg of CH*  emissions per head
               per year, respectively, (based on methane yields  of',6.5 and
               7.5 percent,  respectively).   Given  the distribution  of
               animals, these  values imply  and average of about  58  kg per
               head  per year in  the  U.S.

               Brazil. Argentina.  Australia, and South Africa.   This group
               of countries  has  about  214 million  cattle.  Crutzen  et al.
               examined these  nations  as a  region  because they have well-
               developed beef  industries that rely primarily on  range
               feeding.  Crutzen et  al. therefore  assume that energy intake
               and CH4 emissions for these  cattle  will be similar to the
               estimates derived for range  cattle  in  the U.S.:   feed
               consumption of  110  MJ/head/day, a CH4  yield of 7.5%, and CH4
               emissions of  54 kg/head/year.
                                              /^
               West  Germany.  The  average CH4 emissions  for the  entire West
               German population of  cattle  is estimated  at about 57 kg per
               head  per year.  This  estimate is based on the following:   (1)
               dairy cows -- average intake of 260 MJ/day; 5.5 percent
               methane yield;  94 kg  CH4 emissions  per head per year; (2)
               heifers and steers  between 450 and  600 kg in body weight  --  7
               percent methane yield;  65 kg CH4 emissions per head  per year;
               and (3) 6-24  month  old  animals  -- average intake  of  120
               MJ/day; 6.5 percent methane  yield;  51  kg  CH4 emissions per
               head  per year.

         Based on these estimates  for West  Germany,  the  estimate for the
         U.S. and Canada of 58 kg per year, and the  estimate  of 54 kg per
         year for South Africa,  Argentina,  Australia, and  Brazil, Crutzen
         et al. adopted a value of 55  kg of CH* emissions per head  per year
         as an average for all the developed nations.
     7  This estimate was reported by Crutzen et al.  (1986).   Data in FAO
(1987) suggests that the dairy population in the U.S. has remained at 10
percent of the total cattle herd for several years.  Ensminger (1987) states
that 26.1 million head or 24 percent of total U.S. cattle population are on
feed, leaving 66 percent of total U.S. cattle population on range.  This does
not compare well with the Crutzen et al. values.

-------
                                    A-12
    o    Developing Nations.  Developing nations (excluding Argentina and
         Brazil) have over half the world's population of cattle.  FAO
         reports a 1984 population of 642.7 million.  The nations that
         dominate this population in 1984 include:   India:  195.6 million;
         China 71.2 million; Mexico:  30.5 million; and Bangladesh:  21.9
         million (FAO, 1988).  These nations' populations of cattle account
         for nearly 50 percent of the total in this region.

         Unlike the cattle in developed nations, the cattle in developing
         nations are not used solely for meat and dairy production, but
         also are used as draft animals (Odend'hal, 1972).  These cattle
         are generally fed by-products of the production of food for human
         consumption.

         Crutzen et al. estimated CH4 emissions for cattle in this region
         based on a study of Indian cattle performed by Odend'hal (1972) .
         This study estimated that the daily gross  energy intake of cattle
         in rural West Bengal was 60.3 HJ per head per day.

         To translate the daily gross energy intake into CH4 emissions  per
         head, Crutzen et al. used a CH4 yield of 9 percent from a study on
         nine lactating cows in India (Krishna et al., 1978).  This study
         used face-mask-based gas samples taken once daily to estimate the
         9 percent CH^ yield,  which translates into CHA emissions  of 35 kg
         per year.   Of note is that this 9 percent  estimate is much higher
         than what would be anticipated based on the Blaxter and Clapperton
         relationship.  For feeds of low digestibility  (e.g., 45 to 60
         percent),  the CH4 yield is expected to be  on the  order of 6.5  to
         7.5 percent.  Alternatively, it is at the  low end of the range
         reported by Preston and Leng (1987) (based on stoichiometric
         considerations) for animals on diets that  are deficient in
         nitrogen.   Considerable uncertainty remains regarding the
         contribution of animals in developing countries to global CH4
         emissions.

    The estimates for other animals by Crutzen et al. are based on similar
considerations.   As  shown in Exhibit A-4, Crutzen et al.  estimate emissions
from bovines and other domestic animals to be 73.7  Tg per year ±15 percent.
Emissions from wild ruminants and large herbivores  are estimated to. be 2 to
6 Tg per year.

-------
                                    A-13

    BLAKE

    Blake (1984) presents two estimates of methane emissions from animals
using two different methods:

    o    Estimating CH4 emissions from domestic animals based on studies by
         Ritzman and Benedict (1938); and

    o    Estimating CH4 emissions from all herbivores (including insects) by
         multiplying estimated total dry matter consumption of herbivorous
         animals and insects by the efficiency with which the plant matter
         is converted to methane.

    Domestic Animals.  Blake estimates methane emissions from domestic
animals (cattle, horses, sheep and goats) based on chamber measurements
made by Ritzman and Benedict (1938) on these animals.  Ritzman and
Benedict's findings are summarized below:

    o    Cattle:  Respiration chamber measurements performed on various
         breeds of cattle (average weight of 1400 Ib (635 kg)) showed.an
         average CH4 production rate of 200 g CH4/day/head which was  found
         to be directly proportional to the body weight of the animal.  The
         efficiency of conversion of dry matter intake (by weight) to CH<,
         was 2 percent.

    o    Horses:  Respiration chamber measurements performed on two horses
         (one average sized trotter, weighing 900 Ib (408 kg) and a
         Percheron mare weighing 1500 Ib  (680 kg)) showed that CH4
         production in horses was dependent on the amount of feed intake
         rather than the animal's body weight.  The average CH<, production
         rates for the trotter and the Percheron were found to be 28 and
         106 g CH4/day/head, respectively.  The efficiency of conversion .for
         horses was 0.5 percent.

    o    Sheep and Goats:  Respiration chamber measurements performed on
         seven sheep and three goats showed average CHA production rates of
         15.1 g CH4/day/head for sheep and 14.7 g CH4/day/head for goats.
         The efficiency of conversion for both sheep and goats was 1
         percent.

    For his estimates, Blake adjusts Ritzman and Benedict's data in the
following manner:

    o    Cattle:  Blake asserts that a more representative weight for
         cattle would be 1000 Ib (454 kg) and suggests that even this would
         probably be an upper limit value.  If CH4 production is directly
         proportional to the weight of the animal, a 30 percent decrease  in
         size (from 1400 Ib to 1000 Ib) would result in a decrease in CH4
         production rate from 200 to 140 g CH4/day/head.

-------
                                    A-14

    o    Horses:  The CH4 production rate for horses cited most often from
         Ritzman and Benedict's study  is for the 1500 Ib  (680 kg)  Percheron
         mare  (106  g CH4/day/head).  Blake asserts, however, that a
         Percheron  is 50 percent  larger than the average  horse and consumes
         more  than  twice as much  organic material.  Blake suggests that  the
         900 Ib (408 kg) trotter, with a much lower CH< production rate of
         28 g  CH4/day/head, is a better representative of the average horse.

    o    Sheep and  Goats:  CH4 production rates for sheep and goats is taken
         directly from Ritzman and Benedict's study.

    Blake multiplied population data from the 1970 FAO Production Yearbook
 (1971) by the  adjusted emissions  rates to get yearly world CH< production.
 Blake estimates CH4 production in cattle,  sheep, goats and horses to be 62,
 5.5, 1.5, and  1.2 Tg, respectively.  The sum of these, 71 Tg, is assumed  to
 estimate the lower  limit for CH4 production by herbivores.

    Herbivorous Animals and Insects.   Woodwell  (1970) reports that
 approximately  1.1 x 1017  g/year  dry plant matter is produced on  land,  about
 10 percent of  which is eaten by domestic animals, herbivorous insects
 (i.e., termites) and wild herbivores.  Ehhalt (1974) asserts that an upper
 limit estimate of CH4 production by animals may be determined by assuming
 that all of the 1.1 x 1016  g/year  dry plant  matter  consumed  is converted to
 CHA with the same efficiency as cattle, 2  percent,  to give 220 Tg CH4  per
 year.  Blake asserts that of the  1.1 x 1016  g/year  dry plant matter
 consumed, 43 percent is utilized by cattle.  Therefore, a better estimate
 of CH4 production would assume that 43 percent of the dry plant  matter is
 converted to CH4 with a 2 percent efficiency and the remaining 57 percent is
 converted with a 1  percent efficiency.  These assumptions result in Blake's
 upper limit estimate of CHA production from herbivores of 160 Tg, 95 Tg from
 cattle plus 63 Tg from other herbivores.  Note that Blake's 95 Tg per year
 estimate for cattle using this method  is larger than Blake's 62 Tg estimate
 based on the data from Ritzman and Benedict.
    LERNER ET AL.

    Lerner et al. (1988) present a global database of animal population
densities and associated CHA emissions.   Population statistics were compiled
from FAO Production Yearbooks and other sources.  Published CH4 production
rates were applied to the animal populations to obtain a global
distribution of annual CH4 emissions by animals.

    Animal Population Statistics.  Data on animal populations, in each
country were obtained from FAO Production Yearbooks.  In addition, data on
animal populations for the political subdivisions of the seven largest
countries were obtained from other publications:  Australian Encyclopedia.
Atlas of Australian Resources. Anuario Estatistica do Brasil. Livestock and
Animal Product Statistics (Canada),  Agricultural Statistics of the People's
Republic of China. Agriculture in Brief (India), Agricultural Statistics

-------
                                    A-15

(U.S.), and Europa Year Book (USSR).  Global animal population data
presented by Lerner et al. generally agreed with FAO statistics:  1 percent
or less variation for bovines, buffalo, sheep, pigs, camels and horses; 6
percent for buffalos; and 5 percent for goats.

    Methane Production Rates.  Values for methane production rates for each
type of animal are taken  from Crutzen et al. (1986).  These rates are as
follows:

    o    Cattle:  55 kg CH*/head/year in developed economies;  54 kg
         CH4/head/year in Australia, Brazil, South Africa and Argentina; and
         35 kg CH4/head/year in developing economies.

    o    Sheep:  8 and 5  kg CH4/head/year in developed economies and in
         developing'economies and Australia, respectively.

    o    Pigs:   1.5 and 1.0 kg CH4/head/year in developed and developing
         economies, respectively.

    o    Others:  Production rates  for  camels  (58 kg per head per year),
         water buffalo  (50 kg CH4/head/year), goats  (5 kg CH4/head/year),
         horses  (18 kg CH4/head/year) and caribou (15 kg CH4/head/year) were
         constant for world populations.

    Animal populations were distributed throughout  a 1° latitude by 1°
 longitude grid of the world.  Lerner er al. multiplied the  CH4 production
 rates  by the  populations  in each zone  to get  a global  distribution of CH4
 emissions.  Lerner et al. found  that over 55  percent of emissions from
 animals are concentrated  between 25°N and 55°N latitude.  Global CH4
 production from  domestic  animals is estimated to be 75.8 Tg CH4, 75 percent
 of which is contributed by cattle.

-------
                                    A-16

A. 3 REFERENCES

Baldwin, R.L. and M.J. Allison, "Rumen Metabolism," Journal of Animal
Science. Vol. 57, 1983, pp. 461-477.

Bolle, H.-J., W. Seller, and B. Bolin, "Other Greenhouse Gases and
Aerosols.  Assessing Their Role for Atmospheric Radiative Transfer," The
Greenhouse Effect. Climatic Change, and Ecosystems. B.  Bolin,  B.R. Doos, J.
Jager, and R.A. Warrick, eds., John Wiley & Sons,  New York, 1986, pp. 157-
203.

Blake, D.R., Increasing Concentration of Atmospheric Methane.  1979-1983.
PhD Thesis, University of California, Irvine, 1984, pp. 44-48.

Blaxter, K.L. and J.L. Clapperton, "Prediction of the Amount of Methane
Produced by Ruminants," British Journal of Nutrition. Vol. 19, 1965, pp.
511-522.

Crutzen, P.J., I. Aselmann, and W. Seiler, "Methane Production by Domestic
Animals, Wild Ruminants, Other Herbivorous Fauna,  and Humans," Tellus.  38B,
1986, pp. 271-284.

Ehhalt, D.H., "The Atmospheric Cycle of Methane,"  Tellus. Vol. 26, 1974,
pp. 58-70.

Ensminger, M.E., Beef Cattle Science. The Interstate Printers & Publishers,
Inc.:  Danville, Illinois, 1987.

Ensminger, M.E., The Stockman's Handbook. The Interstate Printers &
Publishers, Inc.:  Danville, Illinois, 1983.

FAO, FAQ Production Yearbook. Food and Agriculture Organization of the
United Nations, Rome, selected years.

Hudson, R.D. and E.I. Reed, The Stratosphere: Present and Future. NASA Ref.
1049, 1979, p. 432.

Hutchinson, G.E., "Circular Casual Systems in Ecology," Annals of New York
Academy of Science.  Vol. 50, 1948, p. 221.

Khalil, M.A.K. and R.A. Rasmussen, "Sources, Sinks, and Seasonal Cycles of
Atmospheric Methane," Journal of Geophysical Research.  Vol. 88, 1983, pp.
5131-5144.

Koyama, T., "Gaseous Metabolism in Lake Sediments  and Paddy Soils and the
Production of Atmospheric Methane and Hydrogen," Journal of Geophysical
Research. Vol. 68, 1963, pp. 3971-3973.

-------
                                     A-17

 Krishna, G.,  M.N. Razdan, and S.N. Ray, "Effect of Nutritional and Seasonal
 Variations on Heat and Methane Production in Bos indicus." Indian Journal
 of Animal Science. Vol. 48, 1978, pp. 366-370.

 Lerner, J., E. Matthews, and I. Fung, Methane Emission from Animals:A
 Global High-Resolution Database. NASA Goddard Space Flight Center,
 Institute for Space Studies, Vol. 2, 1988, pp.  139-156.

 Moe, P.W. and H.F. Tyrrell, "Methane Production in Dairy Cows," Journal of
 Dairy Science. Vol. 62, 1979. pp.1583-1586.

 NRC, Nutrient Requirements of Beef Cattle. National Academy Press,
 Washington, D.C., 1984.

 Odend'hal, S., "Energetics of Indian Cattle in their Environment,"  Human
.Ecology. Vol. 1, No. 1, 1972, pp. 3-22.

 Preston, T.R. and R.A. Leng, Hatching Ruminant Production Systems with
 Available Resources in the Tropics and Sub-tropics, Penambul Books,
 Armidale, New South Wales, Australia, 1987.

 Ritzman, E.G. and F.G. Benedict, "Nutritional Physiology of the Adult
 Ruminant," Carnegie Institute, Washington, D.C. , 1.938, pp. 24-30.

 Rumpler, W.V., D.E. Johnson, and D.B. Bates, "The Effects of High Dietary
 Cation Concentration, on Methanogenesis by Steers Fed Diets With and Without
 lonophores," Journal of Animal Science. Vol. 62, 1986, pp. 1737-1741.

 Seller, W., "Contribution of Biological Processes to the Global Budget of
' CH4 in the Atmosphere," Current Perspectives in Microbial  Ecology. M.J.  Klug
 and C.A. Reddy, eds.,  American Society of Meteorology, 1984, pp. 468-477.

 Sheppard, J.C., H. Westberg, J.F. Hopper, and K. Ganesan;  "Inventory of
 Global Methane Sources and Their Production Rates," Journal of Geophysical
 Research. Vol. 87, 1982, pp. 1305-1312.

 Woodwell, G.  M., "The Energy Cycle of the Biosphere," Scientific American.
 Vol. 223, 1970, pp. 64-74.

-------
                                  APPENDIX B

              IMPACT OF REDUCING LIVESTOCK METHANE EMISSIONS ON
                  GLOBAL WARMING FROM THE GREENHOUSE EFFECT
 B.I  OBJECTIVE

     The objective  of this  analysis  is  to  estimate  the potential benefit of
 reducing methane emissions from ruminant  animals in  terms  of  avoiding
 future global warming.   By reducing these emissions, the concentration of
 methane in the atmosphere  will  not  rise as quickly as it otherwise would,
 thereby delaying,  and possibly  avoiding future global wanning.

     As is commonly done, the  amount of avoided global warming is measured
 in this analysis in terms  of  reductions in equilibrium  temperature change.1
 In this appendix the equilibrium temperature change  is  estimated for  a
 range  of assumptions for the  period 1985  to 2100.

     Of note is that this analysis addresses only the global warming
 benefits of reducing methane  emissions from ruminant animals.  Additional
 benefits of reducing these emissions may  include:

     o      Animal Productivity.   One of the most promising  options for
           reducing methane emissions from ruminant animals is  to increase
           animal productivity,  in particular in developing countries.
           Promising approaches  for  achieving this objective include
           providing supplements  to  animals with important  dietary
           deficiencies  (e.g., non-protein nitrogen supplements for animals
           with diets  that  are nitrogen deficient).  Not only would the
           supplements modify  the animal's rumen fermentation patterns  to
           reduce methane emissions, the increased animal productivity would
           reduce the population of  animals needed to produce a given  amount
           of milk,  meat, or work.   Overall, the improved animal performance
           would be  a substantial benefit.

    o     Animal Wastes.  Animal wastes may be contributing significant
           quantities of methane emissions.  To the extent that this methane
           can be harvested for energy,  the methane emissions will be
           reduced.  The value of the energy derived from the methane  is a
          benefit.  Additionally, carbon dioxide and other emissions  that
     1  Equilibrium temperature change is a measure of the change in the
radiative properties of the atmosphere.  This measure is used to describe
the extent to which the atmosphere is trapping additional energy (like a
greenhouse), thereby warming the earth's surface.  The actual realized
temperature increase that will occur will lag behind the equilibrium
temperature change because it takes time for the feedbacks to occur in
response to changes in the earth's atmosphere and because it takes time for
the oceans to warm and reach an equilibrium with the atmosphere.

-------
                                    B-2

          would have been produced by the energy production that the
          harvested methane replaces will be avoided.2

Given that these benefits may be substantial, it must be recognized that
the analysis that follows includes only a portion of the benefits
associated with reducing methane emissions from ruminant animals.

B.2 APPROACH

    This analysis was conducted with the atmospheric module of the
Atmospheric Stabilization Framework (ASF), recently developed by EPA for
its report "Policy Options for Stabilizing Global Climate," (the EPA
Report).3  For the EPA Report, the EPA developed several scenarios of
potential future emissions of greenhouse gases and other gases that will
influence global climate.  The approach used here is to:

    1.    estimate the impacts on global climate for two of the key
          scenarios developed for the EPA Report (these impacts are
          reported in the EPA Report);

    2.    for each of the two key scenarios estimate reductions in methane
          emissions associated with steps taken to reduce methane emissions
          from ruminant animals;

    3.    create two new emissions scenarios by reducing the methane
          emissions by the amounts estimated in step two;

    4.    for each of the two new scenarios  (with reduced methane
          emissions) estimate the impacts on global climate using the ASF;
          and

    5.    compare the estimated  impacts  from the new scenarios with the
          impacts from the original scenarios  to assess the implications  of
          reducing methane emissions  from ruminant animals.

The differences  in the global warming estimates between the new  scenarios
and the  original scenarios are taken  here as the benefits of reducing
methane  emissions from ruminant  animals.
      2  The burning of the harvested methane produces the same amount of
 carbon dioxide that  the  methane would  have produced had it been  allowed to
 be emitted directly  and  oxidized to carbon dioxide in the atmosphere.
 Consequently,  the burning of the harvested methane does not  add  to the
 carbon dioxide emissions burden.  The  avoided carbon  dioxide emissions  are
 associated with the  burning of coal, oil,  or gas that would  have been
 required to produce  the  energy derived from the harvested methane.

      3  U.S. EPA,  "Policy Options for  Stabilizing Global Climate," Draft
 Report to Congress,  February 1989.

-------
                                     B-3
B.3 SCENARIOS

    To examine future global warming, estimates and assumptions are
required regarding future emissions of greenhouse gases.  These emissions
will necessarily be very uncertain due to uncertainties in the main driving
factors of the future emissions, including:  global economic growth,
population growth, fuel prices, land use, and technological change.
Therefore, the Atmospheric Stabilization Framework (ASF) was developed by
EPA to represent the structural relationships among the key activities that
result in important emissions.  By using plausible estimates and
assumptions in this framework, reasonable scenarios of greenhouse gas
emissions can be produced.

    The two scenarios used here were developed for the EPA Report and are
referred to as the Rapidly Changing World (RCW) and Slowly Changing World
(SOW) scenarios.  These two scenarios provide a range of assumptions about
the future activities that will affect global warming.  The main
assumptions associated with these scenarios include the following:4


          SLOWLY CHANGING WORLD              RAPIDLY  CHANGING WORLD

           Slow  Economic Growth               Rapid Economic Growth

    Continued Rapid Population Growth       Modest Population Growth

     Minimal  Energy  Price Increases       Modest Energy Price Increases

        Slow Technological Change          Rapid Technological Change

        Carbon-in tensive Fuel Mix         Very Carbon-intensive Fuel Mix

         Increased Deforestation              Modest  Deforestation
Of interest for this analysis is that although the SCW scenario has overall
lower projected emissions of key greenhouse gases than the RCW scenario, it
has larger methane emissions from ruminant animals.  Exhibit B-l shows the
methane emissions from the two scenarios, and methane emissions from
ruminant animals.

    As shown in the exhibit, total methane emissions are larger in the RCW
scenario.  The larger emissions are driven primarily by emissions from fuel
production and other energy-related activities.  As also shown in the
exhibit, methane emissions from ruminant animals are quite similar in the
     *  The interested reader is  referred to the full EPA Report for
additional details on these scenarios.

-------
                                     B-4

two scenarios, with the emissions in the SCW scenario being larger in the
later years.  The larger emissions in the SCW scenario are driven by the
larger population growth in that scenario.  A larger human population
requires more food, leading to a larger animal population and larger
methane emissions from animals.  Of note is that these emissions estimates
do not include methane emissions from animal wastes.

    These scenarios imply that methane emissions from ruminant animals may
increase by a factor of 2.4 between 1985 and 2100.  The question is
appropriately asked, is such an increase reasonable, or might there be
important constraints that will prevent such increases from occurring.  The
following observations describe why the scenarios adopted for the analysis
may be reasonable.

    o     Population Growth.  Between 1985 and 2100 the global population
          will grow considerably.  Precise forecasts over 115 years are not
          possible, however.  The range of the population scenarios used in
          the analysis are based on demographic projections by various
          groups and include population growth ranging from a factor of 2.0
          (Rapidly Changing World, ROW) to 2.6 (Slowly Changing World,
          SCW).  If the intensity of animal products produced and consumed
          per capita throughout the world did not change over time, and if
          technology did not change, then in the absence of physical and
          economic constraints, one could expect that methane emissions
          from ruminant animals would increase by a factor of 2.0 to 2.6
          over this period.                  mf,

    o     Economic Growth and Per Capita Consumption of Animal Products.
          The RCW and SCW scenarios reflect a range of future GNP per
          capita growth assumptions.  Although GNP per capita is simulated
          to grow throughout the scenarios (reflecting continued economic
          growth) these growth rates of GNP per capita decline (by
          assumption) over time, and are lower than recent historical
          experience.  Nevertheless, based on existing estimates of the
          anticipated increased demand for animal products that have
          historically been associated with increases in income, and
          including adjustments for cultural factors and increased prices
          that would be expected, the intensity of consumption of animal
          products per capita is generally expected to increase in the
          future.5  This conclusion is particularly appropriate because the
          largest increases in population are anticipated in areas of the
          world that would be expected to increase the consumption of
          animal products per capita with increases in income.  The
          increase in consumption per capita is larger in the RCW scenario
     5  If, in general, the world population moves away from the consumption
of animal products, then the future will be quite different than that
described here.

-------
                                     B-5

           because per capita income  growth  is  larger  in that  scenario.
           Because of anticipated increases  in  consumption per capita,  one
           would expect that methane  emissions  from  ruminant animals  could
           increase faster than the rate  of  growth of  population,  in
           particular in the RCW scenario.

     o     Technology.   Over time one would  expect that  the technology  for
           producing animal products  will change.  In  particular,
           significant advances may be  anticipated in  developing countries.
           In fact,  such changes may  be required  in  order for  the
           anticipated future demand  for  animal products to be met in these
           regions.   The question is  whether such advances will
           significantly reduce the level of methane emissions that may be
           produced per amount of animal  product  produced.  If the ratio  of
           methane per amount of product  produced does not change
           significantly,  then methane  emissions  from  ruminant animals will
           increase due to both population increases and increased demand
           per capita.

           The likelihood of significant  reductions  in methane emissions  per
           amount of product produced,  under the  assumption that steps are
           not taken specifically to  reduce  such  emissions, seems  remote.
           In particular,  if production increases in developing countries,
           the animals  may be expected  to be on forage-type diets,  which  may
           produce more methane per amount of product  than the animals on
           grain-supplemented diets in  the U.S.   If  animals are managed
           intensively  in the future  to increase productivity  in developing
           countries, methane emissions from wastes  could become more
           important if the wastes are  decomposed anaerobically (e.g., in
           lagoons).  Additionally, some  of  the most promising technologies
           for increasing  animal  productivity (e.g., the  use of bovine
           growth hormone  to stimulate milk  production) has only a  modest
           impact on methane emissions per amount of milk produced.
           Although  the number  of animals needed to produce milk may
           decline,  the amount  of methane produced per amount  of milk
           produced  changes  only  slightly.   Finally, if existing techniques
           for promoting animal productivity  are abandoned  (e.g., the use  of
           growth hormones)  then  methane  emissions per amount  of product
           produced  may increase  for a period of time in  some  areas.

    Given  these  three  observations,  the  scenarios of methane  emissions from
ruminant•animals  in the two  scenarios in the EPA Report  seem plausible.
Certainly, alternative  scenarios are also possible,  including scenarios
that presume  significant reductions in methane emissions per amount of
product produced  due to changes  in technology.   Of note  is that efforts are
currently under way  to  identify and evaluate such technologies, so that
they may be factored in to future climate change policy analyses by EPA and
others.  If such  technologies may reasonably be anticipated•to be
implemented quickly over time due to the economic benefits that they
produce in their own right  (without reference to climate change at all).

-------
                                    B-6

then we are indeed in a fortunate position in that the area of ruminant
animal emissions of methane may be expected to decline without incurring
any costs.

    To evaluate the benefits of reducing ruminant methane emissions (in
terms of global warming),  the ruminant methane emissions incorporated into
the two key EPA Report scenarios were reduced by a range of 25 percent to
75 percent.  This range of reduction is taken'from the workshop discussion
and is presented in the findings.  The two "reduced emissions" scenarios
are also shown in Exhibit B-l.
                                                                   /
B.4 RESULTS

    Exhibit B-2 shows the resulting methane concentrations estimated for
the four scenarios.6  Both the ROW and SCW scenarios show substantial
increases in methane concentrations over the current concentrations of
about 1600 ppbv.  Of note is that these increases in concentrations are
influenced not only by the increasing methane emissions themselves, but
also by reductions in concentrations of OH associated with emissions of
carbon monoxide and other compounds.7

    Because reaction with OH is the principal loss mechanism for
atmospheric methane, simulated reductions in OH concentrations result in an
increased atmospheric lifetime for methane.  In the RCV scenario, for
example, the lifetime for methane increases from about 9.6 years (used for
today's lifetime) to about 11.3 years by 2100.  This increased lifetime
means that a given level of methane emissions produces larger atmospheric
methane concentrations than would otherwise be anticipated.  Using the ASF
without this OH-methane lifetime relationship indicates that in the absence
of the OH feedback the atmospheric abundance of methane in 2100 is
simulated to be about 3400 and 2600 ppbv in the RCW and SCU scenarios
respectively, or about 850 ppbv and 500 ppbv lower than the amounts shown
in Exhibit B-2.

    The reductions in methane concentrations associated with the emissions
reduction scenarios are significant, about 200 ppbv to 700 ppbv in each of
the scenarios by 2100.  As noted above, these simulated reductions in
future methane concentrations are influenced by the simulated changes in
future OH levels that also occur.  Without the OH-methane lifetime effect,
the emissions reductions analyzed here would reduce methane levels by
smaller amounts.
     6  Note 1 at the end of this appendix describes the method used to
estimate methane concentrations.

     7  OH concentrations are influenced in the ASF atmospheric module by
the abundances of tropospheric ozone, nitrogen oxides (NOX), methane (CH4),
carbon monoxide  (CO), and emissions of non-methane hydrocarbons  (NMHC).

-------
                                     B-7

    Exhibit B-3 shows the equilibrium temperature change for the scenarios.8
Both the SCW and RCW scenarios result in significant warming by 2100.  The
reductions in methane emissions associated with ruminant animals reduce the
warming by about 0.1"C to 0.4°C by 2100.   Of note is that the avoided
warming is driven not only by reductions in the future atmospheric
abundance of methane, but also by reductions in the anticipated increase in
tropospheric ozone (03).   By reducing methane emissions,  the ASF indicates
that a portion of future increases in 03  can also be avoided.   Because 03  is
a greenhouse gas, part of the warming avoided by reducing methane emissions
is accounted for by this 03 relationship.   The ASF indicates that about 40
percent of the avoided warming is associated with this 03 effect.   However,
this estimate is very uncertain and could be larger or smaller.

    As shown in Exhibit B-4, the avoided warming in the RCW scenario
amounts to about 1.7 to 5.3 percent of the warming by 2050, and about 1.1
to 3.3 percent of the warming by 2100.  In the SCW scenario, the avoided
warming amounts to about 2.0 to 6.3 percent of the warming by 2050 and 2.0
to 6.1 percent by 2100.   The percentages are larger in the SCW scenario
because methane emissions from ruminant animals are a larger portion of the
total anticipated warming in this scenario.

    To put the benefits of these emissions reductions in perspective it is
useful to examine other methods of reducing global warming.  Exhibit B-5
shows estimates of reductions in warming that may be achieved by specific
technological means in the RCW.  As shown in the exhibit, measures for
reducing global warming have relatively small impacts individually.  The .
impacts of reducing methane emissions from ruminant animals presented above
are within the range of impacts achieved by these measures.

B.5 CONCLUSIONS

    This analysis indicates that reducing methane emissions from ruminant
animals will have an impact on future global warming.  Additionally, the
analysis indicates that the benefits achieved from these emissions
reductions are comparable to other measures contemplated for avoiding
future global warming.  Consequently, understanding options for reducing
methane emissions from ruminants should be pursued as part of an overall
investigation into alternatives for reducing future global warming and its
impacts.
     6  These equilibrium temperature change estimates  are  based on a
climate sensitivity of 4°C for a doubling in carbon dioxide concentrations
from pre-industrial levels.  The sensitivity of the earth's climate to
changing atmospheric properties is uncertain, and a range of 1.5 to 4.5eC
has been recognized as an appropriate range of uncertainty.  Recent
analyses indicate that biogeochemical feedbacks may push the range upward.
Note 1 at the end of this appendix presents the method used to estimate the
equilibrium temperature change.

-------
                                      B-8

                                  EXHIBIT  B-l

                        METHANE  EMISSIONS ASSUMPTIONS
        RipUy Changing Wortd


         ToM Cm vrMBom
  T«M CH4 Enwxn ton Aj
                  Anirafc
                                                    Tool CHI EnMm from F
    RCW and  SCW  refer  to  Rapidly Changing World and Slowly Changing World
scenarios, respectively.   The scenarios labeled "with reduction" show a
range of 25  percent  to 75 percent reduction from the baseline rate of
emissions from ruminant animals starting in 1990.  This range of emissions
reductions is shown  for illustrative purposes.  The technical feasibility
and costs of reducing  methane emissions from ruminant animals is currently
being analyzed.
    Source:  Taken from:  U.S.  EPA,  "Policy Options for Stabilizing Global
Climate,"  Draft Report  to  Congress,  February 1989.

-------
                                        B-9   -

                                    EXHIBIT  B-2

                         SIMULATED METHANE CONCENTRATIONS
           RCW Methan* Concentration
                                                            SCW Methane Concentration
4500
4000
3500
3000
2500
2000
1500
       I
             I
                          I
RCW

w/25% reduction


w/75% reduction
                                                 4000
                                                 3500
                                                 3000
                                                 2500
                                                 2000
                                                 1500
                                                 sew

                                                 w/25% reduction


                                                 w/75% reduction
  1985 2000   2025   2050   2075    2100
                                                   1965 2000   2025   2050   2075    2100
      RCW and  SCW refer to Rapidly Changing World and Slowly Changing World
  scenarios, respectively.  The  scenarios labeled "with reduction"  include a
  25 to 75 percent reduction  in  methane emissions associated with ruminant
  animals starting in 1990.   This range of emissions reductions  is  shown for
  illustrative purposes.  The technical feasibility and costs of reducing
  methane emissions from ruminant animals is currently being analyzed.

      Source:  Estimated using the Atmospheric Stabilization Framework
  developed for:   U.S. EPA, "Policy Options for  Stabilizing Global  Climate,"
  Draft Report to Congress, February 1989.

-------
                                    B-10

                                EXHIBIT B-3

                  SIMULATED EQUILIBRIUM TEMPERATURE CHANGE
                                (Degrees C)
     SCENARIO         1985       2000       2025       2050       2075       2100
RAPIDLY CHANGING WORLD SCENARIO
Base Assumptions
25% Reduction
75% Reduction
SLOWLY CHANGING WORLD
Base Assumptions
25% Reduction
75% Reduction
l-A
1.4
1.4
SCENARIO
1.4
1.4
1.4
2.1
2.1
2.0

2.1
2.1
2.0
3.5 5.4
3.4 5.3
3.2 5.1

3.3 4.5
3.2 4.4
3.1 4.2
7.5 9.6
7.4 9.5
7.2 9.3

5.4 6.3
5.1 6.2
5.0 5.9
    The reduction scenarios include a 25 or 75 percent reduction in methane
emissions associated with ruminant animals starting in 1990.  This range of
emissions reductions is shown for illustrative purposes.  The technical
feasibility and costs of reducing methane emissions from ruminant animals
is currently being analyzed.

    These equilibrium temperature change estimates are based on a climate
sensitivity of 4*C for a doubling in carbon dioxide concentrations from
pre-industrial levels.


    Source:  Estimated using the Atmospheric Stabilization Framework
developed for:  U.S. EPA, "Policy Options for Stabilizing Global Climate,"
Draft Report to Congress, February 1989.

-------
                                    B-ll  -

                                EXHIBIT B-4

       PERCENTAGE OF SIMULATED EQUILIBRIUM TEMPERATURE  CHANGE AVOIDED
               BY A RANGE OF REDUCTIONS IN METHANE EMISSIONS
                           FROM RUMINANT ANIMALS
                  ROW
                                                 sew
                      RCW w/ 25% reduction

                      ROW w/ 75% reduction
                                                    SCW w/25% reduction

                                                    SOW w/75% reduction
        2025
2050
2075
2100
2025
                                       2050
2075
21.00
    RCW and SCW refer to Rapidly Changing World and Slowly Changing World
scenarios, respectively.  Percentages  estimated by dividing the reduction
in global equilibrium temperature changes estimated to be associated with
25 to 75 percent reductions  in methane emissions from ruminant animals by
the global equilibrium  temperature changes anticipated for the original  RCW
and SCW scenarios.

    The simulated equilibrium temperature changes displayed in Exhibit B-3
above produce slightly  different estimates of these percentage due to
rounding.

-------
                                          B-12

                                      EXHIBIT B-5

                      REDUCTION IN  GLOBAL WARMING ACHIEVABLE
            THROUGH VARIOUS TECHNOLOGICAL MEANS  IN THE  RCV SCENARIO

                                  ESTIMATES FOR  2050
                       R-onot* Mural (fe*        tepid Rrforwtttlon
                             zsx n»a'n Riaunam*        Nuclear RMV
                                  EnlaBlon* Control        EnlMlora futm
                                       75« RK) 'n Rtnlnantt       Ttranaportatlon
                                  ESTIMATES FOR  2100
                  g    10
                  ^
                  fw

                  fr
                  ?     •
                  I
                  I
                  «     0
                      Prcnot* Mtirai (&•        mpio MTor««tatron
                            2SK MO 'n Runinont*        NUCIMT FOMT
                                  GnlM Centre!         B»l»»te
                                      7SX RM "n RURinents
                                                           Trinaportetron
Note:  Scenarios explained on  the following page.

-------
                                     B-13

                                 EXHIBIT B-5

                    REDUCTION IN GLOBAL WARMING ACHIEVABLE
           THROUGH VARIOUS TECHNOLOGICAL MEANS IN THE RCV SCENARIO

 Explanation of Scenarios:

 1.   Promote Natural Gas:   Assumes  that economic incentives  accelerate
     exploration and production of  natural  gas,  reducing the cost of
     locating and producing natural gas by  an  annual  rate of 0.5% relative
     to the RCW scenario.   Incentives for gas  use for electricity generation
     increases gas share by 5% in 2025 and  10% thereafter.

 2.   25% Reduction in Ruminant Animal Emissions:   25% reduction in ruminant
     animal emissions from  the baseline levels in the RCW.   The baseline  RCW
     emissions do not include emissions associated with  manure  handling.
     Consequently,  assuming that methane emissions from  manure  can be
     reduced,  this estimate is biased downward.

 3.   Emissions Control:  Stringent  NOX and CO controls on mobile and
     stationary sources.

 4.   75% Reduction in Ruminant Animal Emissions:   75% reduction in ruminant
     animal emissions from  the baseline levels in the RCW.   The baseline  RCW
     emissions do not include emissions associated with  manure  handling.
     Consequently,  assuming that methane emissions from  manure  can be
     reduced,  this  estimate is biased downward.

 5.   Reforestation:   Rapid  reforestation so that  the  terrestrial biosphere
     becomes  a net  sink  for C02 by 2000.  This requires reforesting 565xl06
     hectares  (2.1xl06 square miles) by 2015 and I,185xl06 hectares (4,4xl06
     square miles) by 2100.

 6.   Nuclear  Power:   Promote  nuclear  power.  Assumes  that technological
     improvements  in  nuclear  design reduce costs by about 0.5%  per year.

 7.   Emissions  Fees:   Emissions  fees  on fossil fuel production  and
     consumption  in proportion to carbon content.  Production fees of
     $0.50/GJ  for coal, $0.36/GJ  for  oil, and $0.23/GJ for gas  (GJ - giga-
     joule).   Consumption fees of 28%,  20%,  and 13% for coal, oil, and gas,
     respectively.  Production fees implemented fully by 2050, consumption
     fees by 2025.

 8.   Transportation:  Fuel efficiency of new cars in  the U.S. increases to
    40 mpg by 2000.  Global fleet average fuel efficiency reaches 50 mpg by
     2050.

    Source:  U.S. EPA, "Policy Options for Stabilizing Global Climate,"
Draft Report to Congress,  February 1989.

-------
                            NOTE 1 TO APPENDIX B

            METHODS USED TO ESTIMATE METHANE  CONCENTRATIONS AND
                       EQUILIBRIUM TEMPERATURE CHANGE
    The Atmospheric Stabilization Framework (ASF) developed by EPA was used
to estimate future methane concentrations and equilibrium temperature
change presented in this appendix.  The ASF, as well as the bases for the
scenarios used in the ASF, are described more fully in EPA (1989) (the EPA
Report).1  The EPA Report summarizes the working of the atmospheric
composition and temperature module of the ASF as follows:

          The atmospheric composition model was developed for this
    study (Prather, 1988).  It estimates changes in the concentrations
    of key atmospheric constituents and the global radiation balance
    based on the emissions/uptake projected by the other models.
    Perturbations to atmospheric chemistry are Incorporated based on
    first-order (and occasionally second-order) relationships derived
    from more process-based chemical models and observations.  The
    model is essentially zero-dimensional, but it does distinguish
    between the northern hemisphere, southern hemisphere, troposphere,
    and stratosphere.  Global surface temperature change is calculated
    based on the radiative forcing of the greenhouse gases derived from
    Lacis et al. (1981) and Ramanathan et al. <1985) coupled to heat
    uptake by the ocean model using a specified climate sensitivity
    parameter.  This sensitivity parameter is set to yield a global
    equilibrium temperature increase of 2 or 4°C when  C02 concentration
    is doubled, reflecting a central estimate of the range of
    uncertainty .. .*

    This note summarizes the methods used in the ASF to estimate methane
concentrations and equilibrium temperature change associated with methane.
A more detailed description is contained in Prather (1988) and  the EPA
Report.
     1  U.S. EPA, "Policy Options for Stabilizing Global Climate,"  Draft
Report  to  Congress,  February  1989.

     2  The sources cited in  this paragraph include:

     Prather, Michael J., An Assessment Model  for Atmospheric Composition,
NASA Conference  Publication,  1988.

     Lacis,  A.  et al.,  "Greenhouse effect  of trace  gases,  1970-1980,"
Geophysical Research Letters. 8:1035-1038, 1981.

     Ramanathan,  V.  et al.,  "Trace gas trends  and their potential  role  in
 climate change," Journal of Geophysical Research.  90:5547-5566, 1985.

-------
                       Page 2 of Note 1 to Appendix B
    METHANE CONCENTRATIONS

    Methane concentrations are modeled with the following simple annual
integration:
    CONCt - CONCt.j + EMITt -  LOSSt                                        (1)

    where:      CONG - methane concentration (e.g., in ppmv) ;

                EMIT - methane emissions (converted to the units of
                concentrations); and

                LOSS - destruction of atmospheric methane.

The initial concentration (circa 1985 atmosphere) is taken as 1.6 ppmv.  A
variety of emissions scenarios may be used, reflecting varying assumptions
about the known sources of methane emissions.

    The LOSS term reflects the processes that destroy methane in the
atmosphere.  The dominant process is oxidation by the hydroxyl radical
(OH) .   The overall destruction process is modeled using an atmospheric
"lifetime" as follows:

    LOSSt - CONC^ / LIFE,..!                                              (2)

    where:      LIFE - the atmospheric lifetime for methane.

The initial lifetime (circa 1985 atmosphere) for methane is taken as 9.6
years.  This implies that about 10 percent of the methane in the atmosphere
is destroyed by natural processes each year.

    Of note is that the methane lifetime is modeled to change over time  in
response to modeled changes in OH and tropospheric temperature.  The
following equation is used:
    LIFEt - LIFEi *  (1  -  (0.95  *  60H)  -  (0.02  *  «T))                      (3)

    where:      LIFEi - initial lifetime of 9.6  years;

                SOU - calculated perturbation to OH (estimated as the
                average of the northern and southern hemispheric
                perturbations) relative to the reference atmosphere  (as  a
                fraction, e.g., -0.10 is a 10 percent reduction in OH);  and

                61 - perturbation in global average tropospheric
                temperature relative to the reference atmosphere (in
                degrees) .

-------
                       Page 3 of Note 1 to Appendix B
The implication of this lifetime feedback is that as simulated temperatures
change and as OH levels change (due to changes in the abundance of methane,
carbon dioxide, ozone, nitrogen oxides, and non-methane hydrocarbons, and
changes in tropospheric temperature), the lifetime of methane may increase
or decrease.  In the Rapidly Changing World (RCW) scenario presented in the
EPA Report, OH levels are simulated to decline, leading to an increased
lifetime for methane.  However, tropospheric temperatures increase, thereby
reducing the lifetime of methane.  The net effect is that by 2050 in the
RCW scenario the methane lifetime increases by about 14 percent to about
10.9 years; by 2100 the increase is to about 11.3 years.

    These simulated increases in the methane lifetime cause the estimates
of methane concentrations to be larger than they would be in the absence of
the increases in the lifetime.
    EQUILIBRIUM TEMPERATURE CHANGE

    The impacts of increases in methane concentrations are estimated in
terms of an "equilibrium temperature change."  This value describes the
manner in which the composition of the atmosphere has changed the radiative
properties of the atmosphere.  For purposes of evaluating the benefits of
reducing methane emissions from ruminant animals, the anticipated impacts
that lower methane concentrations (due to lower methane emissions) will
have on the radiative properties of the atmosphere must be assessed.  In
addition, however, the impacts that the lower methane emissions and
concentrations have on the abundance of other radiatively important trace
gases must also be evaluated.

    In the ASF, reductions in methane emissions .from ruminant animals not
only reduce the future anticipated methane concentrations, but also have
the following impacts:

    o     Reduce the level of OH reductions anticipated in the future
          (i.e., OH levels are higher than would otherwise-be anticipated).
          Increases in OH tend to reduce the atmospheric lifetimes of other
          trace gases with destruction processes dominated by OH, including
          carbon monoxide (CO), methyl chloroform (MC), non-methane
          hydrocarbons (NMHCs), and partially-halogenated
          chlorofluorocarbons (HCFCs).  These.shorter lifetimes tend to
          reduce their future concentrations slightly.  Because MC and
          HCFCs contribute to global warming, their shorter lifetimes and
          consequent lower concentrations reduce their global warming
          impacts.

    o     Reduce the level of tropospheric ozone (T-03)  increases
          anticipated in the future (i.e. , T-03 levels are lower than would
          otherwise be anticipated).  T-03 is influenced directly by methane

-------
                        Page  4 of Note  1  to Appendix B


           concentrations  in  the  ASF such that  lower future  methane
           concentrations  result  in smaller increases in T-03.  Additionally
           T-03 is influenced  indirectly by methane  concentrations  through
           the modeled impacts of methane concentrations on  CO and column
           ozone,  reinforcing the direct  impacts  of  methane  on T-03,  Because
           T-03 contributes to global warming,  its lower  abundance  reduces
           its global  warming impact.

     The  reductions  in anticipated future global  warming associated with the
 reduced  methane concentrations associated with lower emissions from
 ruminant animals  accounts for the majority of  the global warming benefits.
 However,  the reductions in future T-03 levels  are also significant, thereby
 amplifying considerably the  global warming benefits of  reducing  methane
 emissions from ruminant animals.3  The impact  of the reduced  lifetimes  of MC
 and  the  HCFCs is  simulated to have a small impact on global warming under
 the  scenarios examined in this analysis.*

     The  global warming impacts of methane concentrations are  modeled using
 equations that reflect the "overlapping"  effects that methane has  with
 nitrous  oxide (N20) , and are  as  follows :
    ETCt - (TB / 1.26) * [DRF(CH4t)  -  DRFCCH^) - OVERLAP,.]               (4)

    where:      ETC - equilibrium temperature change due  to methane;

                Ts -  temperature sensitivity of the atmosphere to a doubling
                of carbon dioxide in  degrees C (e.g., 4°C) ;

                DRF() - a function  that estimates the Direct j_adiative
                forcing of a given  methane concentration;

                CH4A - the initial  (i.e.,  pre- industrial) concentration of
                methane (1.02 ppmv) ;  and

                OVERLAPt - the overlapping effect of methane with N20.

Equation (4) is used  each year to estimate the equilibrium temperature
change associated with the concentration of methane in the atmosphere.  The
equation used to estimate the direct  radiative forcing is:
     3  Of note is that T-03,  a component  of urban "smog,"  is  recognized as
a threat to human health, crops, and the environment.  The non-greenhouse
warming benefits of reducing the extent of future increases in T-03 are not
considered here.

     *  The importance of the  MC and HCFC  OH-lifetime feedback could
increase if, for example, the use of these compounds increases
significantly in the future.

-------
                       Page 5 of Note 1 to Appendix B



    DRF - (0.394*CONC°-66 + " 0.16*CONC*e('1-B^OIIC))/(l +0.169*CONC°-62)       (5)

    where:      CONG is the methane concentration in ppmv.

    Because methane and N20 absorb infrared radiation at similar
wavelengths, the use of equation (5) overestimates the implications  of
increases in methane concentrations.  Consequently, the following "overlap"
equation is used to reduce the estimates of the impacts of methane and N20:
                                                                   /
    OVERLAPt - OVL,. -  OVLt                                                (6)
    where:      OVL() - a function that estimates  the  overlap  effect,  once
                with the current concentrations  (subscripted with a "t")
                and once with  the pre- industrial concentrations
                (subscripted with an  "i").

The overlap function is :

    OVL - 0.14*ln(l + 0.636*(CH4*N20)°'75 + 0.007*CH4*(CH4*N20)1-52)       (7)

    where:      CH4 - the methane concentration; and

                N20 — the nitrous oxide concentration.

In interpreting the overlap term in equation (4) ,  one  must keep  in mind
that the overlap  is due to both methane and  N20, and should not be  allocated
entirely to methane.  For purposes of the analysis performed  in  this
appendix, the change in the overlap term caused by the reduction in methane
emissions from ruminants  (and  the subsequent reduction in the  future
increase in methane concentrations) is attributable to the change in the
me thane emi s s ions .

    A separate equation is used to evaluate  the  equilibrium temperature
change of increases in T-03:

    ETCt - (T8 / 1.26)  * 0.00293 * 5(T-03)                                (8)

    where:      ETC - equilibrium temperature change due to T-03;

                T, - temperature sensitivity of the atmosphere to a doubling
                of carbon dioxide in  degrees C (e.g.,  4°C) ; and

                5(T-03) - perturbation to tropospheric  ozone levels relative
                to the reference atmosphere  (circa 1985) in percent.

In the analyses of reductions  in methane emissions from ruminant animals
presented in this appendix, the change in the anticipated future increase
of T-03 associated with the reduced methane  emissions  account  for about 40

-------
                       Page 6 of Note 1 to Appendix B
percent of the benefits in terms of reduced greenhouse impacts.  This
estimate of the T-03 impact is uncertain and could be larger or smaller.
The total impact of the reduction in methane emissions from ruminant
animals is the sum of the change in the equilibrium temperature change
associated directly with methane (equation 4) and associated with T-03
(equation 8).

-------
                       APPENDIX C

"METHANE PRODUCTION BY DOMESTIC ANIMALS,  WILD RUMINANTS.
         OTHER HERBIVOROUS FAUNA, AND HUMANS."

                           by

       Crutzen,  P.J.,  I.  Aselmann,  and W.  Seller


            Reprduced with permission from:

               Tellus.  1986,  pp.  271-284.

-------
Methane  production by  domestic animals, wild ruminants,
                other herbivorous  fauna, and humans
By PAUL J. CRUTZEN. INGO ASELMANN and WOLFGANG SEILER." Max-Planck-lnstuute
     for Chemistry. Depanmenl of Air Chemujry. P.O. Box 3060. D-6500 Maim. F.R. Germany

                  (Manuscript received October 22. 198S: in final form May 30. 1986)
                                        ABSTRACT
        A detailed assessment of global methane production through enteric fermentation by domestic
        animals and humans is presented. Measured relations between feed intake and methane yields
        for animal species are combined with population statistics to deduce a current yearly input of
        methane to the atmosphere of 74 Tg (1 Tg - IO'1 g). with an uncertainty of about 13%. Of
        this, cattle contribute about 74%.  Buffalo* and sheep  each account for 1-9%. and the
        remainder stems from camels, mules and asses, pigs, and horses. Human CH4 production is
        probably leu than I Tg per year. The mean annual increase in CH« emission from domestic
        animals and humans over the past 20 yean has been 0.6 Tg. or 0.75/i per year. Population
        figures on wild ruminants are so uncertain thai calculated CH« emissions from this source may
        range between 2 Tg and 6 Tg per year. Current CH« emission by domestic and wild animals is
        estimated to be about 78 Tg. representing 15-25% of the total CH« released to the atmosphere
        from all sources. The likely CH« production from domestic animals >n 1890 was about 17 Tg,
        so that this source has increased by a factor of 4.4.
          A brief tentative discussion is also given  on the potential CH« production by other
        herbivorous fauna, especially insects. Their total CH« production probably does not exceed 30
        Tg annually.
 1. Introduction

   Methane plays a large role in the photochemis-
 try of the background atmosphere.  It is produced
 and released by various biological  processes and
 is mainly  decomposed in the  troposphere  by
 reaction with hydroxyl radicals (OH). This reac-
 tion  initiates  complicated  reaction pathways
 leading to the formation of various intermediates
 that influence the atmospheric concentrations of
 ozone and  hydroxyl (McConnell  et al.,  1971;
 Levy.-I971; Crulzen. 1973). The hydroxyl radical.
 which is present in the atmosphere at an average
 volume mixing ratio of only 2 x  IO'1*, is primar-
 ily responsible for the breakdown  of natural and
 anthropogenic trace gases in the atmosphere. As
 atmospheric methane has been increasing by over
   * Current address:  Fraunhofer-Institute for Atmo-
  spheric Environmental Research. Kreuzeckbahnstrasse
  19. 0-8100 Garmisch-Partenkirchen. -
1% per year over the past decade (Rasmussen
and  Khalil.  1984;  Blake.  1984;  Seller.  1984).
changes in the background photochemistry of the
atmosphere can be expected. Crutzen (1986) has
postulated that on the whole, global OH concen-
trations are  decreasing (especially outside the
northern mid-latitude zone) and ozone concentra-
tions increasing (especially in northern mid-lati-
tudes and in the upper troposphere).
   Methane  production  by ruminants  has  been
estimated in the past by several authors. The
earliest figures on  world-wide  production  rates
were published by  Hutchmson (1949), who esti-
mated the CH4  emission  by  large herbivores to
be 45  Tg/year   for the  1940's.  Ehhalt (1974)
calculated a global CH«  production  of 100 Tg
CH4 for  1970 from  domestic ruminants. This
 would constitute 20-35% of the  total input of
 methane  to  the atmosphere, which is now esti-
 mated by various authors to be in the range 300-
 500  Tg/year (Khalil  and  Rasmussen,   1983:
 Tellus 38B (1986). 3-4

-------
                          P. J. CRUTZEN. I. ASELMANN AND W. SEILER
Cruuen and Gidel. 1983: Seller. 1984). Crutzen
(1983) and Seiler U984) estimated the global CH«
production from ruminants in  1975 to have been
about 60 Tg and 70-100 Tg. respectively. Similar
values have also been  reported by  Khahl and
Rasmussen  (1983) and Sheppard  et  al.  (1982).
None  of  these  papers,  however,  present  a
thorough analysis of the derivation of these esti-
mates.  In  this paper,  we  will  give  a  detailed
account of the worldwide methane production by
domestic  livestock  and wild ruminants,  and by
humans, using the extensive  information which
has  now  become   available   on  methane
production by animal species.

2. Energy utilization  and methane  production
    in animals

   The  energy content  in food is transformed in
 the process of digestion and partly lost as chemi-
 cal compounds in faeces, urine and fermentation
 gases.  The rest  is  used  to produce  heat. to.
 perform  body work or to build new body tissue
 (Fig.  1).  The magnitude of the various losses of
 energy depends on the animal species and on the
kind and  quality of the  feedstuff. Measures of
feed quality are digestibility and metaboltzabtlin,
which are the fractions of the gross energy intake
that are  convened to digestible  energy  and
metabolizable energy, respectively.
  In   adult  homeotherms.  basal  metabolio*
defines the  minimum  energy  demand under
conditions of thermoneutrality and total rest. The
daily basal metabolism, expressed in megajouiet
(MJ).  is roughly proportional to the A power of
the  body weight W (kg) and is given by the
formula:
 basal metabolism - 0.293 W*".
(1)
   f
-------
                               METHANE PRODUCTION BY ANIMALS
                                         273
hones and the elephant snowea CH1 yields in the
rangeot  1.5°;-3°;.
  Blaxter and  Clapperton (19651. in analysing
numerous data on methane production by cows
and sheep  have shown  that  at  maintenance.
methane yields increased from 7 3°; to 9*£ when
the digestibility 01" the feed was raised from 65°;
to 95*i- The methane yields (6.5-7°;) were inde-
pendent of the digestibility at a feeding level of 2
 x maintenance and decreased from 6°; to 5°; at
3  x maintenance  when  the  feed  digestibility
changed from  60°.; to 90*;. Similar  results were
reported  by Van der Honing et al. (1981) who
observed methane yields  on  5-6.5°; for dairy
cows fed on 3.1-3.5  x maintenance.  Wainman et
al. (1978) measured CH, yields of 7.9*.{ from
steers fed on  1.5  x  maintenance  and  a 77*;
digestible diet. Krishna et al. (1978) estimated
higher CH4 yields of 9% in Indian cattle fed on a
slightly above maintenance diet and low quality
feed.
   Published methane yields from sheep show a
somewhat larger ranee of values. Murray et al.
(1978)  found CH,  yields to  rise from 3.5TJ to
 5.6" with increasing feed intake in  Merino ewes
 fed  on  protein-rich  lucerne  chaff  with  a
 digestibility of 63";. Kemp ton and Lcng (1979)
 observed methane yields  of 5.4-6.4°; in growing
 lambs, being highest on low protein diets, while
 Seclcy ct al. (1969) measured 8.2-97°; in adult
 sheep ted on ryccrass hay at maintenance level.
   Non-ruminating.  * ell-nourished  pigs  have
 much  lower methane  yields  (0.4-0.9°;) which
 seem  10  be  independent   of  feed  mixture
 tSchneiaer and Menke. 1982). but pigs given low
 quality feeds may be expected to show CH* yields
 between  1°.  and  2%  (H.  Stemgass. personal
 communication).
   From these data, it becomes apparent that the
 calculation of the  global methane  release  from
 domestic animals requires consideration of differ-
 ent animal management ana  feeding schemes.
  Published  information is scarce. We have there-
  tore also made  considerable use of information
  "hich  was  kindly  provided to us by  experts
  ihrougn personal communications.


  3. Methane production by cattle

    In ihe L'S. there are 5 types of cuttle: cuiry
  -owv ceet cuttle on ie;c. ana cuttle on runge. In
Europe and in the USSR, range cattle are rare.
ana (he rearing of dual-purpose cattle for dairy
-na beef production is common practice. In other
Countries like Australia or Argentina, cattle are
kept  primarily on  range and dairy cows are
relatively less numerous.
  •Because  cattle  are fed at different levels of
energy intake and  quality of food  in different
pans of the world,  we will next present  some
detailed analyses for the US. West Germany and
India. From these data, we will extrapolate to
other pans of the  world  to arrive  at  global
estimates of methane production.
   The average number of feed units (I feed unit is
the digestible energy contained  in  I  pound of
corn) consumed in the US by milk cows (includ-
ing heifers), cattle on feed, and  cattle on range
are  101 SO.  6650 and 4800 per animal per year.
respectively (G. Allen, personal communication).
With an energy content of 6.57 M J per feed unit.
this convens to 230 MJ. 150  MJ and 110 MJ of
daily gross energy intake, which corresponds to
about 2-3. 1.75 and 1.3  x  maintenance.  Using
 Blaxter and Clappenon's C.965) methane yields
of 5.5'i. 6.5*; and  7.5°,; of gross energy intake
 for  these   categories,  and  an  energy  content
 of  55.65   MJ/kg  CH4. the annual  methane
 production per individual in the aforementioned
 categories arc 84 kg. 65 kg ana 54 kg. respective-
 ly. The  US cattle population  is made up  of 10"
 dairy cows. 12.5°; beef cattle on.feed ana 77.5°;
 cattle on range (Food and Agricultural Organiza-
 tion of the United Nations  (FAO). 1984).  The
 mean methane production  by cattle in the US is
 there tore equal to 58 kg per animal per year.
   In   West  German  statistics   (Statistisches
 Bundesamt.  1984). cattle  are grouped  in age
 classes: younger than 6 months (2.7 millioni. 6 to
 24 months (6.2 million) and older than 24 months
 (6.5 million). The latter are subdivided into dairy
 cows (5.4 million), and  heifers or  steers (l.l
 million).
   The  feeding level for dairy- cows  in highly
 productive stages  is  3 to 3.5  x  maintenance.
 .-quivalent to 300-320 MJ  per day (e.g..  Van aer
 Honing  et al..  1981. Kirchgessner. 1985).  but
 mean gross energy  intake over a year's period, is
 around 260 MJ per day (H. Stemgass. personal
 communication). With  a methane yield of 5.5°^,
 •AC calculate a \early CHA production of 95 kg per
 oairv cow.
       JaB M9S6). 3-i

-------
                          P. J. CRUTZEN. I. ASELMANN AND W. SEILER
  For honors and steers with body weights of 450
kg ana 600 kg. mean gross energy intakes are
125  MJ  and   165  MJ  per  animal  per  day
(Kirchgcssner. 1985). respectively, of which 7%
is lost as methane (e.g.. Wamman et al., 1978).
This converts to a yearly methane production  of
65 kg per animal as a weighted mean for heifers
and steers.
  Within the age class younger than 24 months.
about 10*4 of the animal population is kept for
veal production. These animals are fed liquid.
milky food which strongly limits CH* production.
Calves younger than 6 months are  also fed pri-
marily on milk and other highly, digestible feed
which do not promote methane production.
  Adopting a  mean body weight of 340 kg in the
age class 6-24 months, the gross energy intake is
about 120 MJ per animal per day (Kirchgessner.
 1985). As the  quality of the feed changes during
animal growth from highly digestible to mixtures
 including more roughage, we adopt a mean CH4
yield of 6.5%. Consequently,  we estimate a
 production of 51  kg of methane per animal per
 year in this ag? class.
   Combining the  derived  methane yields with
 the cattle populations in the given age classes in
 West Germany,  an  average, annual  methane
 production of 57 kg per head is calculated.  As
 other European countries  have relatively  more
 beef than dairy cattle and thus slightly  lower feed
 intakes (H. Steingass. personal communication).
 we will adopt a mean annual  CH4  production of
 55 kg per bovin for the developed world by  extra-
 polation of the data derived for the US and West
 Germany. This figure may have an uncertainty
 factor of about 1ST;, mainly due  to  remaining
 uncertainties  in the methane yields.
   Similar methane production rates apply  for S.
 Africa.  Australia. Argentina and  Brazil.  These
 countries have cattle which are kept  mainly on
 range and fed on roughage (D. E. Johnson.
 personal communication). For these countries, we
 assume  the CrL production rate to be equivalent
 to that  of US cattle  on range, i.e..  54 kg  per
 animal  per year.
    Despite  higher  methane  yields,  individual
  methane  production  rates  from  cattle  in  the
  developing world are lower, because feed  intake
  is near to. or only slightly above, maintenance.
  consisting mostly of roughage or kitchen  refuse.
  Pandev (1981) determined the average daily gross
energy intake  for grazing cattle in  Varanasi
(India) to be equal to 52.5 MJ  for cows. 19 Mj
for calves and  73.5 MJ for bullocks. Odend'hal
(1972) estimated  the mean daily gross energy
intake in a  cattle  population of 3800 individuals
in a  West  Bengal district to  be 60.3  MJ per
animal. In this  population. 14% of the cattle were
dairy cows. This is close  to the mean  for all of
Asia (15.6%) and Africa (12.6%) according to the
FAO (1984). Adopting an average feed consump-
tion of 60.3 MJ for cattle in the developing world
and  a CH4 yield of  9% for  low quality  feed
(Krishna et al..  1978. H. F.  Tyrrell,  personal
communication),  we calculate  a mean annual
CH4  production rate of 35 kg per animal in the
developing  world.
   According to the FAO  (1984). the world cattle
population  in  1983 was 1.2 x  10". Of this. 33%
was  kept  in  the developing  and 47%  in the
developed  world. Brazil,  and Argentina. Adopt-
ing the annual methane production rates of 35 kg
and  S3 kg. respectively,  for these  regions, we
calculate a global mean annual CH4 production
of about  45  kg  per head of  cattle,  which is
considerably less than the 73  kg  adopted by
 Ehhalt (1974). The global methane release to the
 atmosphere from cattle  totals 54 Tg annually
 (Table I).  Of  this about 40% is produced in the
 developing world.
 4. Methane  production  by  other  domestic
    ruminants

   Buffalo; are kept in some Third World coun-
 tries for milk production and  for  farm work.
 Body weight and feed demand are higher than for
 cattle,  so that  buffalos  require a  gross energy
 intake  of 85  MJ per animal per day (Pandev.
 1981).  We again assume a CH* yield of 9% and
 therefore estimate a production of 50 kg CH* per
 animal per year. Given a total population of 124
 million,  the production of  CH.,  from  buffalo*
 equals about 6.2 Tg per year.
    Adult sheep,  which weigh 60-70  kg. have  a
 gross  energy  diet  of 30-40  MJ  per  day  in
 Germany and the US. while immatures are fed
 20-25  MJ daily (Kirchgessner.  1985: National
  Academy of Science. 1975). In West Germany.
  more  than 40% of the  sheep  population is less
  than 1-year-old, so that a gross energy intake of
                                                                          Teilus 38B (1986).

-------
                              METHANE PRODUCTION nv \W =.!_:>
prutlticnoii or tiamesne ununais inm ii:in:i:n\ /.••  -V.-' (!
                                                                    I0': 5)
.Animal type
:nd regions
Cattle
Developed countries. Brazil
and Argentina
Developing countries
Total
Buffalo*
Sheep
Developed countnes
Developing countnes and
Australia
Total
Coats
Camels
Pits
Developed countries
Developing countries
Total
Horses
Mules. Asses
Humans
Population
lx 10»|
572.6
652.3
1225.4
124.1
399.7
737.6
1137.3
476.1
17.0
32S.S
444.S
773.6
64.2
53.9
4669.7
C'rt. prout:ct:on CH, production Production
rv.T tnuiviuu.il by total population grand total
Ufi-vearl iTg-'year| |Tg/yearl
:: 1-1.5
:-5 ::.s
54.3
:0 6.2
S 3.2
5 3.7
6.9
5 2.4 2.4
JS 1.0 1.0
1.5 0.5
1.0 0.4
0.9
13 1.2 1.2
10 05 0.5
0.05 0.2 0.3
Total
                                                                  73.7
23 MJ per day be taken as an average. Murray et
at. (1978) give smaller values of 15-19 MJ per day
Tor adult Menno ewes in Australia, which may be
partly due to a smaller body weight of about 40
kg. Similar  values are given by Mathers and
Wallers (1982)  Tor England.  Adopting a  mean
daily energy intake of 20  MJ in  the developed
countnes  and 13 MJ  in  Australia and  in  the
developing world and applying a mean CH* yield
of 6°; (Kempton and Leng. 1979; Murray et al..
 197S). we calculate production rates of 8 and 5 kg
CH4 per animal per year for the developed  and
less  developed  world, respectively.  The  world
sheep population in  1983  wms estimated by the
                                 FAO  (1984)  to  be  1.1 x  10».  about equally
                                 divided between the developed countries and the
                                 developing  world,  including  Australia.  Sheep
                                 therefore produce about 6.9 Tg of CH4 annually.
                                 worldwide.
                                   The individual, average gross energy intake of
                                 goats in India was measured by Pandey (1981) to
                                 be 14  MJ  per  day.  This leads to a methane
                                 production  of about 3  kg per goat  per year.
                                 similar  to  that for  sheep.  The  world goat
                                 population of about 476 million produces  a total
                                 of 2.4 Tg methane per year.
                                   Data on feed intake by camels were not avail-
                                 able but may be estimated from their average
Tel-as 5SB (1986). 3-

-------
                          P. J. CRLTZEN. I. ASELMANN AND \v  SEILLR
body weight 01 570 kg  (Nowak and  Paradise.
1983). Applying the general formula (1) for basal
metabolism, we calculate a  mean  minimum en-
ergy demand of 34 MJ per day.  The ratio of the
mean basal metabolism of cattle in India with a
mean  body weight between 200-350 kg  (H.
Steingass. personal  communication)  to the  re-
ported gross energy intake is about 1:3. Applying
this ratio to camels, we calculate a mean gross
energy intake  of  100  MJ per animal per day.
Camels  live essentially on  roughage.  With 9?,
methane yield as in the case of Indian cattle, we
calculate the methane production from camels to
be 58 kg per individual per year, which results in
a global yearly CH* emission of 1 Tg from a total
of 17 million  camels. Other  camclids (Llamas.
Alpaca,  etc.)  have   populations   too  small
(McDowell, 1976) to produce  globally significant
amounts of CH*.
 5. Methane  production  by  non-ruminant.
   domestic animals
        methane  yield  from  pigs  on  highly
digestible fattening feeds is less than  I " of the
gross energy intake (e.g.. Schneider and Mcnkc.
1982). Taking data from  Europe, gross  energy
intake is between 12.5 MJ  per day for young pigs
and about 90 MJ  per  day for  lactattng sows.
Based  on  age   and  weight-class  population
statistics from West  Germany,  we calculate  a
mean individual gross energy intake of 38  MJ per
day. Of this, about 0.6" is released as methane
(Schneider and  Menke.  1982). yielding  1.5 kg
CH4  per year.  We  assume  that this number
applies to developed countries.  In developing
countries, the animals are smaller and less well
nourished, with diets consisting commonly of
kitchen refuse or green fodder. With these foods.
methane yields might reach  2% (H. Steingass.
personal communication). Assuming the gross
energy intake to be } of  that in the developed
world and adopting a methane yield of 1.3°;, we
calculate a yearly CH* production rate of 1.0 kg
in the developing world. Multiplied with the pig
populations in  the  developed and developing
world, this yields about  I  Tg CHA per year from
 pigs.
   Methane yields for horses are between those
 for pigs and ruminants. They equal 3-*°, of the
digestible  energy  or  2-3",  of  the gross enem
intake (Kirchsiessncr. 1985). The energy demand
for norses \viih ;i  mean  body weight of 550 k«.
executing  mcuium work loads  for 2 h a day. is
about 73 MJ of digestible energy or 110 MJ of
gross energy (National  Academy  of  Sciences.
1973). Similar values are given by Kirchgessner
(1985).  If 2.5°e of the  gross  energy  intake is
released as methane, we calculate a mean yearly
CH., production of 18 kg per animal. The world
population of 64 million horses therefore  pro-
duces a total of 1.2 Tg CH4 per  year. Similarly,
we deduce a mean production  rate of 10 kg per
animal per year for the  global  mule and donkey*
population of 54 million, leading to a total emis-
sion of 0.5 Tg of CH4 per year.
 6.  Methane production by humans

   Mcthanogenic bacteria in the large intestine of
 humans produce amounts of methane which vary
 greatly between individuals. The percentage of
 healthy humans  who  produce methane  ranges
 from about 30" to more than  S0*£ (Bond et al.,
 1971; Djorneklctt  and Jensen.  1982: McKay et
 al.. 1985). The  methane produced  in the  large
 intestine is partly absorbed by the blood within
 the colon  wall and exhaled through the lungs, and
 partly excreted in flatus gas.                  ;
   Measured CH4 mixing  ratios  in exhaled  air
 from methane-producing individuals vary widely.
 from only a few ppm above ambient air ratios to
 more than 70 ppm. with an average of 14.8 ppm
 from 280  healthy individuals (Bond et al.. 1971).
 Levitt and Bond (1970) reported a mean  value of
 21 ppm CH*. In a series of expenments on 120
 healthy humans. Bjorneklett and Jensen (1982)
 observed  a  medium methane mixing  ratio in
 exhaled air of  16.3 ppm.  This  leads  to mean
 individual CHA exhalation rates of 40-50 g per
 year, assuming  a mean breathing  volume of "
 l.mm (Schulz. 1972). The methane exhalation by
 the human population of 4.7 x 10" is therefore.
 equal to about 0.2 Tg per year. This surprisingly
 small quantity is totally negligible in the global
 CH4 budget.
    Kirk (1949) measured 2-8°; methane in  the
 flatus gases from a group of 20 individuals. The
 average production of flatus gas from this group
 was 1.5 mi. mm. Extrapolating  this information to
                                                                         Tcllus JSB (19861. .:~

-------
                               METHANE PRODUCTION DY ANIMALS
(he global  human population. ;hc  rcie.iso  01'
methane in flatus gas is estimate to be about 0 I
Tg/year.  These  small   meinane  production
numbers  arc  in agreement  with  other  siuuics
(Sieggerda.   1968:  Marthr;sen and  Fleming.
1932)"
7. Methane release from wild ruminants  and
   other large herbivores

  The  global  protection of  CH*  from  wild
ruminants is difficult to estimate due to lack of
sufficient data  or animal populations and feed
intake.  Some   population   assessments  exist.
however, for ce-tain regions of the world and can
be extrapolated to  global conditions. McDowell
(1976) gives •- population figure of 27 million for
wild ruminants in the northern temperate regions
(except China). These ruminants are comprised
mostly of deer and moose. In Table 2. we have
listed  their  mean  body size  (Nowack  and
Paradiso.  1983) and feed intake (e.g.  Nystrom.
1980:  Sadleir.   1982). As wild  ruminants  live
entirely on roughage and herbs near maintenance
lew.s."we assume a CH4 yield of 9%. Using these
figures,  we obtain  a total release of 0.4 Tg of
r ethane by wild  ruminants in the temperate
 egions. mostly from deer.
  Information  on  populations and  mean  body
weights of wild ruminants  in the Serengeti  is
summarized in Table 3 (Houston. 1979: Western.
1979). The total population  of about 2 million
mainly consists of gazelle and wildebeest. Data
on  gross energy intake  in  Table 3  have been
calculated from formula (I)  for the  basal meta-
bolic rate, multiplied by a factor of 2 to give the
                         iikcly gross  energy  requirement of free  living
                         ungulates  (Mocn.  1973:  Ehringham.   1974).
                         Again. 9*o of the gross energy intake is assumed
                         10 be released as methane. With this information.
                         the   OH*  production  in  the  Serengeti   from
                         ruminants is estimated  to be about 0.02 Tg per
                         year.  Assuming the CH* production in the
                         Serengeti  to   be  representative  of   global
                         conditions, the total CH4 production by the wild
                         ruminant population of 100-500 million  in the
                         subtropical  and  tropical  regions (McDowell.
                         1976) may be estimated to be of 1-5 Tg per year.
                         Together with the contribution from ruminants in
                         the  northern temperate  regions, the annual CH*
                         production from  wild  ruminants in  the  world
                         may. therefore, be equal to 2-6 Tg per year which
                         is small compared  to  the  CH« production by
                         domestic animals.
                           Statistics  on methane  production  by   other
                         large, non-ruminating herbivores in the Serengeti
                         are  likewise listed in Table 4. The most important
                         contributions come  from zebras and elephants.
                         The total methane production is less than I0?i of
                         that  by  the   ruminants.   Altogether,   non-
                         ruminating  large herbivores  are  a  negligible
                         source of atmospheric methane.
                         8. Methane emission by other fauna

                           The consumption of plant matter by the large.
                         wild herbivores considered so far. sums up to 50-
                         200 Tg dry matter. The average consumption of
                         plant matter by herbivorous fauna is estimated to
                         be 1% of the net primary productivity (NPP) of
                         natural ecosystems, or about 7000 Tg dry matter
                         (Whittaker.  1975). Consequently, relatively small
 Table 2. CHt production by wild ruminants in temperate northern regions


Species
moose, elk
white and black-tailed
deer, mule deer, red
deer, reindeer, caribou
roe deer

Populations
(* 10»J
8.3


::o
40
Mean body
weicht
(kg)
350


90
15
Gross energy
intake
(MJ/day)
S3


:6
5
CH, production
per individual
(kg year)
31


15
j
CH4 production
total •
(Ttyear)
0.03


0.33
0.01
 Total
26S
0.37
TeJIus J&B (19S6). 3-»

-------
rs
                          P J. CRCTZEN. 1. \SELMANN AND W SE1LER
Table 3. CHt pwaucnon or wild amniais m the

Species
wildebeest
buffalo
Thompson's fazelle
lirarTe
eland
topi
impala
kontoni
waterbuck
Grant's gazelle
Total ruminants
zebra
elephant
wanhof
hippopotamus
rhinoceros
Total non-ruminants
Population
(x I0^|
72000
10300
98100
1700
2400
5600
11900
2100
300
600

240
5
34
i
1

Mean oooy
weicnt
(kgf
123
450
15
750
340
100
40
125
160
40

:oo
1725
45
1000
S20

Gross energy
intake
(MJ/tfayl
^
57
4
34
46
19
9
i^
:6
9

31
157
10
104
90

CH. production
per individual
Ike. year!
13
\4
i
50
27
11
5
13
15
5
*
5
26
1
17
IS

CH. production
total
(I0»kt/year|
9.4
\ 7
J. *
2.0
O.I
0.6
0.6
0.6
0.3
O.I

IS.l
1.2
0.13
0.03
0.03
0.02
1.4
 Table 4. Trends in tlomesitc animal populations and C/14 production rates

Cattle
population (10"!
CH« production rate (kt-anima! year)
CH. proouction total |Tg yearj
Sheep
population (10*)
CH. proouction rate (kf, -animal year)
CH. proouction total {Tg.yearj
1S90

•10
:-s
11

590
^
*
1921 -15

640
33.3
:s

645
5.4
3
1941-45

:•«>
J0.5
30

7S5
5.6
4
1961-65

1016
415
t *
• j

1007
5.9
6
19S5

j«-e -.
^ ^


11""
1 1 •
<,
O
 Other domestic animals, humans etc.
   CH4 production total (T|.year|

 Total CH* proouction (Tg,year|
                             II
                                                                             60
 CH * yields from other fauna than that considered
 so far couid  produce  substantial  amounts of
 methane.  In the following, we  will briefly con-
 sider the possibilities.
   According to Whittaker H975K the consump-
 tion oi* punt matter by fauna is 10-15°; of  the
 NPP in grasslands and 4-8% in  forest and wooc-
 lands. However, only a tew investigations on the
 energy transfer through the food webs in these
 ecosystems have been published so far. which
give information on the main herbivorous ccs-
iumers.  Norton-Griffiths  U979)  and  Sincur
(1975) estimated the consumption of plant mar.sr
by small mammals, mainly voles, in the Serenssti
to be !.:•; of  the above-ground  NPP in  lcr.i-
grass savanna  and  only O.l»;   in  shon-grJi
savanna. Much more. 7.6°; and  4.1%, resce:-.-
ively, is consumed by invertebrates (mainly grii»-
hoppers), in these two grassland sites
1983).
                                                                                JSB(19S6>. .--

-------
                               METHANE PRODUCTION BY ANIMALS
                                                                                        :*9
  In the palm savanna at Lamto. Ivory Coast.
the main consumers are fungus-growing termites
ib", of NPP). followed by grasshoppers tO.40;)
ma small  rodents, mainly voles (0.22%). Large
herbivorous   ungulates  play   no  major  role
(Lamotte and Bouriiere. 1983). Certain ants ana
caterpillars are also important consumer* but
quantitative information  on their role  is not
available.
  In the Fete Ole savanna in Senegal, termites
have also  been found to  be major harvesters of
plant tissue, consuming 10*; of the NPP (Josens.
198?). Gillon  (1983) states that at this site, grass-
hoppers and caterpillars may consume as mucn as
grazing mammals,  but  gives  no data on  their
plant consumption.
  In  undisturbed  temperate  grassland   sites.
primary consumption by invertebrate herbivores
is 0.5-9*.  of the above ground NPP. being lowest
in poor, unproductive sites  and increasing with
productivity   (Andrzejewska  and  Gyllenberg.
 1980).  Wiegcn and Evans (1967) determined an
upper  limit of 12!;  for  plant consumption  by
herbivores in an  old  grass field site in South
Carolina,  mainly  by invertebrates. Only I "t was
consumed by vertebrates, mainly field mice and
savanna sparrows.
   In tropical  forests. Janzen  (1983) considers
 defoliating animals as the most important plant
 consumers,  mainly  moth  larvae,  caterpillars.
 beetles and  other invertebrates.  Jordan (1983)
 cites data, indicating that usually no more than
 .*•-«*, of (he  leaf  biomass is harvested by insects.
 In nutnent deficient, low-production forests, the
 consumption may drop to 1*.',.
   Herbivorous vertebrates are not important as
 primary  consumers  in  tropical  forests.  Owen
 (1983) gnes  a very low figure of only 72 kg/knv
 for the density of herbivorous mammals at the
 Tano  Nimn  Forest (Ghana), consisting of 3
 ungulates and 7  pnmates. These low figures are
 also confirmed for South Eastern  tropical ram
 torests (Whitmor. 1984). In  contrast, in some
 South American forests, larger  populations of
 siotns and tapirs  may be t'ound that may consume
 j considerable fraction of plant production with-
 m the torests (Janzen.  1933).  Rodents  are  also
 present in tropical forests  but apart from some
 population  figures, no  energy  intake data are
 available.
    In temperate torests.  primary consumption by
               . 3-4
insects is at most 10-1S", of the above-ground
NPP  (Remmert.  1980).  Franklin (1970)  cues
:ome ricures tor broad-leaved trees, which range
trom 3-8"„. However, these data contradict those
t'rom an intensively investigated forest site in the
North-east US. in which Gosz et al. (1978) found
less than i % of the total  NPP. including roots, to
be consumed by herbivores. In approximate or-
der 01 importance,  these are chipmunks, mice.
foliage-eating insects, birds, deer and hares.
  Consumption  figures  from  boreal forests or
tundra sues are not available. Remmert (1980)
gives a rough figure of 1-2*.; for consumption by
vertebrates at Spitzbergen (Norway), but gives no
estimates for invertebrates.
  Summarizing the above widely different data.
it appears that major consumers in  natural habi-
tats are  invertebrates, mainly  insects,  and to  a
much lesser degree small herbivores. If we as-
sume that  small herbivores (mainly rodents and
lagomorphs. i.e.. hares, rabbits, etc.) in natural
habitats consume  less  than  1% of the  global
 NPP. with a CH« yield of 1.5?; (Johnson, person-
 al communication), we calculate for this group  of
 animals an upper limit of methane  production  of
 2-3*; Tg per year.
   An  estimate  for  methane  production   by
 invertebrates is  even more speculative. So far.
 methane production  has only  been measured  in
 termites (Zimmerman  et  al..  1982. Rasmusien
 and Khahl. 1983: Seller ct al.. 1984. Fraser et Jl..
  1986)  and in  wood-bonng   larvae  of  beetles
 (Bayon.  1980)  which  utilize  symbiotic  micro-
 organisms in their digestive tract to break down
 cellulose. According to Swift et al. (1979).  most
  insects  harbour flagellates in their  guts for cellu-
  lose digestion. Thus it is likely that insects and
  probably other  invertebrate primary consumers
  also harbour methanogenic micro-organisms.
    Measurements from different genera  of ter-
  mites indicate methane yields of less than 0.01*,
  to  1.5*; (Zimmerman  et al.. 1982:  Khaltl and
  Rasmussen. 1983: Seller et al.. 1984). If we acoot
  this range for all invertebrate primary consumers.
  the  consumption of 6°; of the global NPP by
  herbivorous invertebrate probably couid proouce
  at  most IS Tg annually.  However,  this upper
  limit estimate is speculative as long as no data on
  possible CH4 production yields by insects other
  than termites are available.  According to Owen
  (1983). plant-feeding  insects in tropical  forests

-------
280
                          P  J CMTTZEX. I. ASELMANN-AND W SEILER
are  much  more  abundant  than  commonly
assumed, due 10 inadequate counting metnoas.
Consequently, more work on the potential CH*
production by invertebrates is certainly justified.
  Our upper estimate of IS Tg for inverteorate
primary consumers overlaps to some degree with
global production estimates for termites (includ-
ing herbivorous species, wood- and dung feeders.
soil feeders and other J). which range from 2-5 Tg
(Seiler et al., 1984) up to ISO Tg (Zimmerman et
al.. 1982). The most recent study by Fraser et al.
(1986) suggests an annual production of 6-42 Tg
with an average of maybe 14 Tg.
   So far we have primarily been concerned with
primary consumers. Most of the plant  matenal
produced each year by vegetation is lost to the
litter layer and subsequently  decomposed by  a
large variety of organisms. A survey through the
literature on  soil biology has given no hint on
methane production in soils apart from anaerobic
environments and termite mounds. Swift et al.
(1979) report that certain cockroaches and dung
 beetles  rely  upon symbiotic   bacteria and
 protozoans  for  the   digestion  of   structural
 nolysaccharidcs.  This  could  imply   possible
 methane  production as in the case of  termites.
 Generally, methane can  most  prooably only be
 produced  by macrulauna consumers of plant de-
 tritus (primary saprotrophsl as these are the only
 ones with true anaerobic  digestive tracts. Organ-
 isms belonging to  mesaiauna  are probably too
 >mall in size to develop anaerobic conditions in
 their guts iGrewe. personal communication). The
 macrofauna  plays a major role in the tropics and
 declines  in  importance  towards the  poles. In
 colder climates,  major  breakdown of organic
 compounds  is accomplished by microfauna and
 fungi (Swift et al..  19*9). which implies a gradi-
 ent of decreasing  methane production  potential
 from soil fauna from the tropics to the poles.
   Earthworms have attained special interest  in
 the past  for their ameliorating effect  on soils.
 These  animals  feea on  detritus together --nth
 mineral panicles and attain huge turnover rates
 of matter within me soil. Their guts produce a
 favourable environment for a diverse population
 of micro-organisms. However, cellulose  is  \ery
 badly  digested  (Bruuns. '.96S>  .ina   methano-
  genesis is not reporter. Measurements on acrocic
  Mills of  temperate anu  tropical  grassland  sues
i Seller et al.. 1984. 1986) all indicate methane
decomposition in  the soil and no production.  If
soil-dwelling  organisms  produce  methane,  it
seems likely that this is readily decomposed with-
in the soil and does not escape to the atmosphere.

9. Current and  past CH4 production  from
   domestic animals

  Data  on  the  estimated  global   methane
production by domestic animals and humans  in
1983 are summarized  in Table I. Total methane
release  to  the atmosphere is about 74  Tg/year
with an estimated uncertainty of about 13% By
far the largest contribution, about  54 Tg. comes
from cattle. About 40*; of this emission occurs in
the developing world. Next in importance come
buffalo and sheep, which produce about 6 Tg and
7 Tg  CH4 per  year, respectively.  Goats and
camels produce  2.4  Tg CH«  and  I  Tg CH4
annually. Non-ruminant, domestic animals each
year emu about 2.6 Tg CH« to the atmosphere.
The  human  contribution is only  about 0.2  Tg
CH4  per year. In comparison with  the estimated.
global methane emission of 74 Tg by domestic
animals in the  year  1983.  the input  by wild
ruminants of 2 6 Tg is relatively small.
   Because of the growing  world  population  of
humans and  its growing  food  demand,  (he
population of domestic  animals   has  also  in- ;
creased considerably  during the  last  century.
 leading to  an   increase  in  global  methane
 production rates. According to data published by
 Mulhall (13921 for the end of the last century and
 data  for  the  tint  half of  this  century  (US
 Department  of  Agriculture.  1936-1970).  the
 world cattle population has increased  from  310
 million in 1390 to 640 million in 1920-1925. and
 "40  million in  1940-1945. The  corresponding
 figures for sheep are 590 million. 645 million, and
 "55 million, respectively (Table 4i According to
 the available statistics, the sheep and cattle popu-
 lations show similar  trends for tne period 1920-
  1960. but  large differences for 1390-1920. which
 is probably  muicative  of less reliable  statistical
 information from the early period.
    More ucuiled statistical information is avail-
  able lor the time period after 1940. particularly
  due  to the  -.vork by the Food and Agricultural
  Organization 01 the United Nations iFAO. 1973.
                                                                           Tcihn 'S

-------
                               METHANE PRODCCT1ON DY ANIMALS
                                         IS I
1-J82 and 1934). According to these data, during
;:-.e last two uecudes. the world's cattle ana sheep
populations have grown by O.S", and  06°, per
scar, respectively,  reaching  figures in 1983 01
1225 x 10" for caitie and 1137  x 10" for sheen.
Similar  annual increases  in  global population
numbers are round for pigs (1.4"). burTalos 11 "„).
goats (1.2°;) and camels (0.5*;). The population
01 mules and  asses stayed  about constant, while
that  of horses showed a slight decline of 0.25°0
per  year.  According to the  available  statistics.
since I960, the growth rates of the cattle  and
bheep populations have slightly declined.
  The temporal trends of global CH* emission by
ruminants are not only dependent on their popu-
lations but also on quality and quantity of  feed
•.make. In the developed countries, the average.
individual feed intake  has increased during the
last  20 to 30 yean, although the  corresponding
increase  in CH4 emission may have been com-
pensated to some extent by declining CH4 yields
due  to  higher  iced  quality.  In the developing
world, both feed intake and quality  may have
declined.
  To estimate individual  methane production
rates prior to  1983.  we assume  that in  1890.
average methane yields in the world were about
(.•qual to current yields in  the developing world.
 A rule between  1X90 and 1980. MO adopt a linear
 .•row in  m individual CII* emission rates. With
 ::us aviumnuon. the CHt release trom cattle tu»
 *nore man ouadrupled during the la»i century
 ;:om II  Ti  in 1890  to  54 Tg  m  1983.  CH,
 emission by »hecp  has grown from 3 Tg to  7 Tg
 Curing the same period (Table 4). Information on
 uecltning populations  of wild  ruminants is not
 a\ailable. but they can certainly  not  have  been
 >uiricicm to  compensate  substantially for the
 Drouth  in   methane  emissions  by  domestic
 .intmalv Altogether, our analysis  indicates that
 '.no  total annual CH4 emission from all animals.
 including wild ruminants, has increased from 2!
 Ta m 1890 to aoout 78 Tg in 1983.

 10. Conclusions

   Methane is produced by the anaerobic fermen-
 'aiion oi organic matter in the rumen and lower
 jut Di  domestic and wild  animals.  The  CH,
          rite rer animal is dependent on quant;.
and amount ot feed intake. Based  on data that
.:re presently available, we estimate  that about 5-
•>"., of the gross energy intake by ruminants is lost
10 methane production. Lower methane yields m
the ranee u.5-5"0 are derived for other domestic
animals such as pigs, horses, etc.
  The current global CHt emission by domestic
.md wild animals is estimated to be 73 Tg/year.
From this, almost  80*;, or about 60 Tg/year.
comes irom cattle and  bulTalos. The  rest is pro-
duced by sheep (7 Tg-yearl. wild ruminants (2-6
Tg yean, and others. About 40% of the total CH,
produced by domestic animals is emitted in the
developing countries, mostly in Asia, followed by
South  America  and Africa.  The  main source
region for methane  from cattle in the developed
world is North America (11 Tg year) followed by
Europe (8 Tg yean and the USSR (7 Tg  year).
   Methane  production by domestic and wild
animals constitutes about  15-22*; of the total
troposphenc CH4 input,  recently  estimated by
various authors  to  be  in  the  range  of  300-500
Tg/year (Khalil  and Rasmussen. 1983:  Crutzcn
and  Cidel. 1983: Seiler.  1984). Production by
animals  represents  one of- the most important
individual  sources within the  troposphenc CH»
cycle.  It  is about  two times larger than the
production  trom coal mining  and  natural gas
!cak> tSeiier. 1984. Crutzcn. 1936: Dollc et al..
 1986). The unly  emissions which could  be larger
are those irnm  the anaerooic cccay 01 organic
matter  in  rice  fields and   natural  wetlanus.
 Methane release  from rice paddies may  be aoout
70-130  Tg year, if information  obtained   in
 Italian rice rieids can be extrapolated to tropical
conditions i Holzaplel-Pschorn and Seller. 19861
   The total emission  of  CH4  by  domestic and
wild  animals has increased from about  21  Tg  in
 1890 to 4ft T* in 1940 and 73 Tg in 1933. maimy
due to crowing populations of cattle. burTalos and
iheep According to these figures,  the mean rate
of increase m CH, emission by domestic and wild
 animals curing  the last 43 \ears has been I  1*,
 per \ ear.
   We ha\e estimated a tentative  upper limit  ot
 about ••) Tg CH.,  emission per >ear.

-------
                            P  1. CRCTZEN. I. ASEUMANN AND W SEILER
11. Acknowledgements

  Dr. G. Allen (US Department 01' Agriculture.
Economic Research Service.  Washington DC).
Dr. H. F. Tyrrell (US Department of Agriculture.
Science and Education Administration. Beltsville
Agricultural  Research Center.   Beltsville).  Dr.
D.  E.   Johnson  (Michigan  State   University.
Department of Animal Husbandry. East Lansing.
Michigan.. Dipl. rer. nat. f. Grewe. Insmut fur
Angewandte   Bodenbiologie.   Hamourg.  and
especially  Dr.   H.  Stemgass  (Umversuat  of
Hohenheim.  Institut  fiir Tierernahninf, Stutt-
gart. F.R.G.) supplied much information through
private communications. This work was partly
supported by the  Ministry  for  Research  and
Technology of the Federal Republic of Germany
through grant BMFT KBF 68.
                                           REFERENCES
 Andnejewska.  L. and GyHenbert.  G. 1910.  Small
   herbivore subsystems. In: Grasslands', systems anaivsu
   and man. eds.: A. I. Breymeyer and G. M. Van Dyne.
   IBP 19. Cambridge University Press. Cambridge, p.
   201-268.
 Bayon. L.  I9SO.  Volatile fatty  acids and  methane
   production  in relation to anaerobic carbohydrate
   fermentation  in ORYCTES  NASICORNIS larvae
   (Coleptera: Scarabalidael. J.  liuect fhynoi. 26. 819-
   828.
 Bjorneklcti.  A. and Jenssen.  E.  1982. Relationship
   between   hydrogen  (H;)   and   methane  (CH,)
   production in man. Gastroenierototv 17 (8). 983-992.
 Blake. O.  R. 1984. Increasing concentrations of atmo-
   spheric  methane.  1979-1983. Ph.O. Dissertation in
   Chemistry. University of California. Irvine. 213 pp.
 Blatter. K. L. and Clappenon. J. L. I96S. Prediction 01
   the amount 01 methane produced by ruminants, Br
   J. .Vmr 79.511 -522.
 Blaxter. K. L. and Oerkawski. J. 1966. Modifications
   on  the  methane  production  of  the  sheep  bv
   supplementation 01 its diet. J. So. Fd. Agric. 17.417-
   421.
  Bolle. H.  J..  Seller. W  and Bolm.  B.  1986. Other
    greenhouse  gases and  aerosols. Assessing their role
    for  atmospheric  radiative transfer.  In:  it'UO'
    ICSL'.'L'XEP International Assessment of the Kite ol
    carbon dioxide ami other radianteiv atttrt constituents
    m climate rananon and associated impacts, in press.
  Bond. J.  H.. Engel. R  R. and Levitt. M.  D.  I9*M.
    Factors inrluencini pulmonary methane excretion in
    man. J. Exa. Med  133. 572-588.
  Bournere. F.  1983.  The savanna mammals:  Introduc-
    tion. In: Ecosystems at the world 13. Tropical savan-
    nas, ed. F.  Bourliere.  Elsevier.  Amsterdam—Ov
    lord—New York, o J«:°-*6l
   3rauns. A. 1968. Prsmiscne Bodenbtotot.it. G. Fiscner.
    Stuttgart. 4~0 pp (in German).
   Crutzen. P. J. 1986. The roie of the tropics in atmo-
     >pnenc chenustrv  In: (jtoonvsioiaty 01 Amazonia, cu.
     R. Dickinson. Wi'.ev. New York im nresst.
   C.-uizen.   P   J.   !'*^3.   Mmosonenc  interactions—
      -.omoccneou* *js reactions 01  C. N. jr.c S contain:r.j
              s. In:  Ti.t miiuir niotetter.emicat i. \ctcs _T_J
   iheir interactions (Scope 21). eds. B.  Bolm and R. B.
   Cook. John Wiley. Chichestcr. p. 69-II2.
 Crutzen. P. J. 1973. A discussion of the chemistry of
   some  minor constituents  in  me stratosphere  and
   troposphere. Pure Appt. Grophvt.  106-109.  1385-
   1399.
 Crutzen. P. J. and Gidel. L. T. 1983. A two-dimension-
   al photochemical model of the atmosphere. 2:  The
   iroposphenc   budgets  of   the    anthropogenic
   chlorocarbons. CO. CH«, CH,G and the effect of
   various  NO,  sources  on  troposphenc  ozone. J.
   Ceophyt. Res. M. 6641-6661.
  Ehhalu O. H. 1974. The atmosphenc cycle of methane.
    7Y//IU 26. 58-69
  Eltnngham. S. K. 1974. Changes in the large mammal
   community of  the  Mweya  Peninsula.  Rwcnton
    National Park.  Uganda, following the removal of
    hippopotamus. J  Appl. Ecol.  II. 855-856.
  Food and Agricultural Organisation tFAO) 1973.  1982.
    1984.  Production yearbook  1972. Vol.  24  Rome.
    Production  yearbook  '1981.  Vol.   35.  Rome.
    Production yearbook 1983. Vol. 37. Rome.
  Franklin.  R. T. 1970.  Insect influences  on the forest
    canopy. In: Analvsa ol temperate toresi ecosvsiems.
    Ecological Studies I. ed.  D. E. Reichle. Spnnger.
    Berlin—Heidelberg—New York. p. 86-99.
  Fraser. P. J..  Rasmussen.  R.  A..  Crerfield. J. W .
    French. J. R. and KhahL M. A.  K. 1986. Termites
    and  global methane—another assessment. J Atmos
    Chem. in press.
  Gillon. Y. 1983. The  invertebrates  ol the grass layer.
     In: Ecosvstems ol the *orld 13. Tropical wannas. ed.
     F. Bourliere. Elsevier. Amsterdam—Oxford—New-
     York, p. 289-311.
   Gosz.  J. R.. Holmes. R. T . Likens. G  E. and Borman.
     F. H. 19*3 The no* 01 energy in a torest ecosystem.
     Set. American. 233 «3). 92-102.
   Holzaptel-Pichom. A. and Setter.  \V  1986. Methane
     emissions dunng a veeetattve period from an  Italian
     rice paodv. J. Cmonvs Rti. m press.
   Houston. D. C.  '.979  The acactattons of scavengers.
      In: iVrrnfeii. Dynamics at an *eos:ittm. eos. A. R. E.
      Sinclair  anU  M.   Nonon-GnrTuh.  University cf
      Chicaco Press. Chicago— Loncon. :p. 253-136.
                                                                                   Tcilu*;-SB(l986). j-

-------
                                   METHANE PRODUCTION BY ANIMALS
                                                                                                  133
Hutchinson. G. E. 1949  A note on two aspects ot (he
  ccocnemistry of carbon. Am. J. Sci. 247.17-32.
Janzen. D. H. 1983.  Food »eos: Who eats what. »h\
  how and with what effects in a tropical forest.' In:
  Ecotviiems  of  the world  14.4,  Tropical ram  lorest
  tcos\ stems. Structure and tunetton. ed. F B. Gollev
  Elsevier. Amsterdam—Oxford—New York. p. 167-
  182.
Jordan. C. F. 1983.  Productivity of tropical rain forest
  ecosystems  and the  implications for their  use  as
  future wood and energy sources.  In: Ecosvstemi oi the
  world 14A.  Tropical ram lorni ecoirttems.  Sirurturt
  and function, ed. F. B. Gollev. Elsevier. Amsterdam—
  Oxford—New York. p. I1T-I36.
Josens. G. 1983. The soil founi of tropical savannas III.
  The termites. In: Ecoivttemt 01 the world 13, Tropical
  latarutas. ed. F.  Bouriiere. Elsevier. Amsterdam—
  Oxford—New York. p. 105-5:4.
Kempton. T. J. and  Leng. R. A. 1979. Protein nutrition
  oi frowmf lambs. 1. Responses  in growth and rumen
  function  to  supplementation   of  a  low-protein-
  cellulosic   diet   with  either   urea,  casein   or
  formaldehyde-treated  casein.  Br. J.  Nutr. 42. 239-
  ?02.
Khahl.  M. A. K. and  Rasmussen. R. 1983. Sources.
  sinks, and seasonal cycles ot atmosphenc methane. J.
  Cenphn. Ret. M.  3131-5144
Kirchgcssncr. M.  1985.  T^rrmaknm^.  DLG-Verlag
   Frankfun. 6. Auflage. 481 pp (in German).
Kirk. E. 1949. The quantity and composition of human
   flatus. GauroeHierology I? (S). 782-794.
 Kleibcr. M.  1961.  The lire ot  lift. An introduction in
   jnimul t-nrrtrrirj. J.  Wiley and Sons.  New York —
   London, p. 277-?22.
 Krishna. G..  Razdan.  M. N   and Rav. S.  N. 1978
   Effect oi nutritional and seasonal variations on heat
   and methane production in Bos mama,  Indian J.
   Anon. Sci. 49. 366-370.
 Lamotte. M. and Bouriiere. F  1983.  Energy flow  and
   nutnent cycling in tropical Savannas. In: Ecosrstrms
   til the *orld 13.  Tropical tarannas.  ed. F  Bouriiere.
   Elsevier.  Amsterdam—Oxford—New  York. p. 583-
   603.
 Levitt. M. D. and  Bond. J. H. 1970. Volume, compo-
   sition, and source of intestinal gas. G»istroenteroior.\
    <9. 921-929.
 Levy. II. H.  19~1.  Normal atmosphere: large radical
   and  formaldehyde concentrations predicted. Satnce
    173. UI-I4J.
 Lieth.  H. 19*5.  Measurement of calonnc  values. In:
    Pnniarv proauctwitv oi :r.t biaipnere.  eos.  H. L.ctn
    and   R.  H.  Whittaker.  Springer.  New  York—
    Heidelberg—Berlin,  p. 119-129.
  Mannmsen. O. and  Fleming.  S. E.  1982. Excretion of
    breath and flatus gases  =v  humans consuming high-
    hbre oiets. J. \uir. I/.'. 11 }3-l 143.
  Mathers. J  C. ana Walters.  D. E.  1982. Variation in
    -r.etnane procuction bv s.-.s;a  lea every two hours. J
     lew. 5
Slavnard. L. A and Loosli. J. K. 1962. Amniat nutrition
  McGraw-Hill.  New  York— Toronto— Lonaon.  p.
  :--»n->«)i
McConneil. J. C.. McElroy. M. B. and Wofsv. S. C.
  197)  Natural  sources of  atmosphenc CO. .\>tiurt
  ::J. 187-188.
McDowell. R. E. 1976. Importance ot rumuiaiut oi the
  Hur/J tor non-loud utfi. Cornell International Atncul-
  ture Mimeograph. 52. 35pp.
McKav. L. F . Eastwood. M. A. and Brydon. W  G.
  1935. Methane excretion in Man—a study of breath.
  ilatus. and faeces. Cut 26. 69-74.
Menke.  K.-H. and  Huss. \V 1975.  Titremohna*  und
  Fuih-rmHulkuiide. UTB-L'lmer. Stuttian. JI9pp. on
  German i.
Moen. A. N   1973. U'ilJIift tcohfy: art analytical ap-
  proach.  Freeman.  San Francisco, p. 109-131.
Mulhall.  M.  G.  1892. The  dictionan- of ttatunei. G.
   Routledge and Sons. London. 108-110.
Murray. R*. M..  Bryant. A. M. and  Leng.  R. \. 1978.
   Methane production in the rumen and tower gut of
   sheep given lucerne chaff: effect  of level of intake.
   Br. J. Star. J9. 337-345.
National Academy of Sciences. 1975. Nutrient require*
   menu of sheep. Ntaneni rtqwrmt*t ot donetiKtttetl
   anunai*. no. 3. 5th revised edition. 42-45.
 National Academy  of Sciences. 1973. Nutnent require-
   ments of horses.  Ntannn reowtmnt a)  dontnttcaifd
   anunait. no. 6. 3rd revised edition  if-19.
 Nonon-Gritliths. M.  1979. The  ir.iuence of gracing.
   browsing, and fire on the vegetation dynamics oi the
   Screngeti.  In: iVrrnrrn.  D\nr.-nicj ol an «TOJIt/rm.
   cUs.  A  R. G.  Sinclair an>   M.  Nonon-Gnthths.
   tm>crmv of Chicaeo Pre  1981. Vegetation and bovine population
    interar o-s   in  i!:e   >avanna  grazmguncs  oi
    Chanarsarabha sanctuarv-. Varanasi.   II. Seasonal
    behaviour ot grazing animals and an  assessment of
    earn TIB capacity oi the arazmtlanas.  Tropical Ecol.
    ::. ro-i$6.
  Rasrnjssen. R. A. and  Khalil. SI. A.  1C. 1934. Atmo-
    »p-«nc  methane   in   the  recent   and  ancient
    a mosohercs:    concentrations.    trends.     and
     nternemisphenc  tradient.  J.   Gtopnvs.  Ra.  39.
    11599-11605.

-------
                             P. 1. CRLTZEN. I. ASELMANN AND W SEILER
Rasmusscn. R. A. and Khalil. M  A  K.  1983. Global
  production ot metnane by termites. .Vuiure 101. "00-
  701
Remmen. H. 1980. Ecology.  A  ttxttook. Sprinter.
  Berlin—Heidelberg— New York. 2S9pp.
Ritzman. E. C. and Benedict. F  G. 1938. \utntianal
  pHvaotofr e) ins adult ruminant. Carnegie Institution.
  Washington O.C.. p. 14-M).
Sadlctr.  R.  M. F. S. 1982. Energy consumption and
  subsequent panitiomni in lactatina Diack-tailed deer.
  Cm. J. Zaat. 60. 382-386.
Schneider.   W    and   Menke.   K.   H.   1982.
   Untersuchungen  uber den enertetischen Futtcrwen
   von Metasse-Schnitzein in Rationen iiir Schweme. Z.
   TurpMiiotof rr.  Titremahrun^  und Futiermiiitlkundt
   •HI. 233-2401 in German).
 SchuU.    E.   1972.    Phvuoitajie.    Urban    and
   Schwarzenberg. Mtinchcn—Berlin—Wien. p. 84 (in
   German I.
 Seeley.  R. C.. Armstrong.  O. G. and MacRae. J. C.
   1969.  Feed carbohydrates—the contribution of the
   end-products of their digestion to energy supply in
   the ruminants. In: £nrrfy mnabottun of farm anmmls.
   Proc.  4th Symp.  Warsaw.. Poland. Sept. 1967. cds.
   K. L. Blatter. J. Kiclanowski. G. Thotbek. Onel
   Press. Newcastle upon T>-ne. p. 93-102.
  S«iler. W.  19X4. Contribution ot  biological processes to
   the global budget of  CH.  in the atmosphere. In:
    Current  nmnfftnrs in mierohut.ttotuty. eds. M. J.
    Kluc and C. A. Rcddy. Amencan Society for Micro-
    biology. Washington. DC. LSA. p. 468-JT7
  Scilcr.  W . Conrad. R.  and Schanfe.  O. 1984  Field
   'studies oi methane emission from termite nests into
    the atmosphere  und  measurements 01 metnane urn
    uke by tropical soils. J. Aimet. Clu-m. I. 171-186.
  Seller.  W .  Conrad. R.. Scharrfe. O. and  Giehl. H.
     1986. CH, uptake by soils. Free. 2nd. tnirmaiionai
     Simpouw en BuMpnerf^iimospntrt txcnanjt. Mainz.
     March 16-21. I9S6.
  Sheppard. J. C.. Westberg. H . Hopper. J. F.. Ganesan.
     K. and  Zimmerman.  P.  1982. Inventory 01  global
     methane sources  and  their production  rates.  J.
     Gtopkn.Ra.t~ i:-OS-l312.
Sinclair.  A.  R. E.  I97S. The  resource limitation  of
  trophic levels in  tropical grassland  ecosystems.  J.
  .4mm. Ecal. 44. 497-520.
Statisiisches Bunaesamt. 1984.  Statitiuchn Jahrbuek
  Jer Bumittrtpublik DeuiKhiand. Kohlhammer. Slutt-
  gart. !58-l59(in Germin».
Steggerda. F.  R.  1968.  Gastrointestinal gas following
  food consumption. Aim. N. Y. Acad. Set. ISO. 37-66.
Swift. M. J.. Heal. 0. W. and Anderson. J. M. 1979.
   Dtcomnuino*  at  terminal  ecosvittnu.  Siuditt ui
   Ecoioir •!• University of California Press. Berkeley-
   Los Angeles. 372 pp.
 U.S. Department of Agriculture. 1936-1970. Agrmatur-
   al statistics. Washington DC. USA.
 Van der Honing. Y.. Wiemann. B. J.. A. Sieg and von
   Donselaar. B. 1981. The effect of fat supplementation
   of concentrates on digestion and utilization of energy
   by productive dairy cows. AfcrA. J. Afne. Set. 29. 79-
   92.
  Wainman. F. W.. Smith. J. S. and Oewey. P. J. S. 1978.
   The  predicted and  observed metabolizabk energy
    values of mixtures of maize silage and barky fed to
   cattle. Rowett Research Institute. Aberdeen.  Scot-
    land. 55-58.
  Western. 0.  1979.  Size,  life history and ecology in
    mammals. A/r. J. Ecoi. 17.  185-204
  Whitmor. T- C: 1984. Tronteal ram tufrjtsof the far east.
    Clarendon Press. Oxford. 2nd. edition. 322 pp.
  Whittsker. R.  H. I97S. CoHumunwt and teosntfms
     McMillan Publ. New York. 2nd. edition, p. 385.
  Wicccrt. R. G. and Evans. F C. 1967. Investigations 01"
     Secondary productivity in grasslands. In: 6VroM«irr
     [•roftucttrnv ol  terminal (rosftifmt. tol.  II. Cd. K.
     Petrusewicz. Panstwowe  Wydawmctwo  Nauko»c.
     Warsazwa—Krakowie. p.  499-MX.
   \Vohn. M.  J.  1981.  Fermentation  in the  rumen ana
     human Urge intestine. Scitnee .'/J. 1463-1461.
   Zimmerman. P.  R.. Greenberg.  J. P.. Wandiga. S. O.
     and Crutzen. P. J.  1982. Termite: a potentially large
     source ot atmospheric methane, carbon dioxide ana
      molecular hydrogen. Sonnet 218. 563-565.
                                                                                     Teilus.-.SBll986). 3-i

-------
  APPENDIX D




WORKSHOP AGENDA

-------
 METHANE  EMISSIONS  FROM RUMINANTS

                 FEBRUARY  27-28, 1989
            PALM SPRINGS,  CALIFORNIA
                        WORKSHOP AGENDA
SESSION 1: FEBRUARY 27, 1989
LOCATION:  GARDEN ROOM, SPA HOTEL AND MINERAL SPRINGS
8:30am

9:00am - 9:15am



9:15am - 9:30am

9:30am - 9:45am

9:45am • 10:00am


10:00am -  10:30am



10:30am -  10:50am


10:50am -  ll:10am

llrlOam -  12:15pm

12:15pm -  l:00pm

l:00pm - l:15pm


l:15pm - 2:00pm
Coffee, juice, and breakfast rolls

Welcome and Announcements, Michael Gibbs, ICF

INTRODUCTION AND BACKGROUND

      Workshop Objectives, John S. Hoffman, EPA

      Methane in the Atmosphere, Don Blake, U.C Irvine

      Overview of Methane from Ruminants and Key Workshop
      Questions, Michael Gibbs, ICF

      Discussion

SOURCES OF METHANE EMISSIONS

      Indirect Calorimetry Measurements, Don Johnson, Colorado
      State University

      Rumen Processes and Models, Lee Baldwin, U.C Davis

      Discussion

      Lunch Brought In

      Waste Fermentation as a Source of Methane,
      L.M. (Mac) Safley, North Carolina State University

      Discussion
February 23, 1989

-------
                      WORKSHOP AGENDA (Continued)
SESSION 1: FEBRUARY 27, 1989 (continued)
LOCATION:  GARDEN ROOM, SPA HOTEL AND MINERAL SPRINGS
2:00pm - 2:20pm


2:20pm - 2:40pm


2:40pm - 3:15pm

3:15pm  3:45pm


3:45pm - 430pm

5:45pm
CHARACTERIZATION OF POPULATIONS AND EMISSIONS

       Demography and Ecology of Cattle in Developing Countries,
       Stewart Odend'hal, University of Georgia

       Dependence on Cattle Populations in Developing Countries, Jim
       Ellis, Colorado State University

       Discussion

       Systems Approach to Evaluating Methane Emissions, Jim Fadel,
       U.C. Davis

       Discussion

MEET IN LOBBY FOR BUS TO DINNER ON MT. SAN JACINTO
 February 23, 1989

-------
                       WORKSHOP AGENDA (Continued)
 SESSION 2: FEBRUARY 28, 1989
 LOCATION:  GARDEN ROOM, SPA HOTEL AND MINERAL SPRINGS
 8:30am

 9:00am • 9:15am
 9:15am - 9:30am

 9:45am - 10:00am

 10:00am - 10:15am


 10:15am  12:00

 12:00 - 12:45pm




 12:45pm - l:00pm

 l:00pm - 3:00pm
Coffee, juice, and breakfast rolls

Announcements, Michael Gibbs, ICF

IDENTIFYING AND EVALUATING OPTIONS FOR REDUCING
METHANE EMISSIONS FROM RUMINANTS

       Overview, Michael Gibbs, ICF

       Microbial Ecology of the Rumen, Bob Hespell, USDA

       Productivity Enhancement and Methane Emissions, Henry
       Tyrell, USDA

       Discussion

       Lunch Brought In

NEXT STEPS FOR REDUCING METHANE EMISSIONS FROM
RUMINANTS

       ARS Planning and Priorities, Lewis Smith, ARS

       Discussion of:

                   Projects for improving emissions estimates

                   Projects for evaluating emissions reductions
                   options

                   Forming and Ad Hoc Group

             -     Others to involve
Workshop Adjourned
February 23, 1989

-------
        APPENDIX E




LIST OF WORKSHOP ATTENDEES

-------
  METHANE EMISSIONS  FROM RUMINANTS

                   FEBRUARY 27-28,  1989
             PALM  SPRINGS,  CALIFORNIA
                        WORKSHOP ATTENDEES
Curtis Bailey
CSRS
USDA
Room 330
Aerospace Building
Washington, D.C. 20251-2200
(202) 382-1534

Lee Baldwin
Department of Animal Science
University of California, Davis
Davis, CA 95616
(916) 752-1250

Donald Blake
Department of Chemistry
University of California
Irvine, CA 92717
(714) 856-5011

Lauretta Burke
U.S. EPA
Room 3009 PM 220
401 M Street, SW
Washington, D.C 20460
(202) 475-6632

Dan Dudek
Environmental Defense Fund
257 Park Avenue South
New York, NY 10010
(212) 505-2100

Jim Ellis
Natural Resources Ecology Lab
Colorado State University
Fort Collins, CO 80523
(303) 491-1643
Jim Fadel
Department of Animal Science
University of California, Davis
Davis, CA  95616
(916) 752-1250

Michael Gibbs
1CF Consulting Associates, Inc.
10 Universal City Plaza
Suite 2400
Universal City, CA 91608-1097
(818) 509-7150

Keith Gomes
California Milk Producers
11709 E. Artesia Blvd
Artesia, CA 90701
(213) 865-1291

Thomas J. Goreau
324 North Bedford Road
Chappaqua, NY  10514
(914) 238-8768

Champ Gross
CALF News
7252 Rammet Ave
Suite 205B
Canoga Park, CA 91303
(818) 346-2068

Juan Guererro
UC Cooperative Extension Service
Courthouse
938 9th Street
El Centra, CA 92243
(619) 339-4250

-------
                             WORKSHOP ATTENDEES
                                        (Continued)
Robert Hespell
USDA Northern Regional Research Center
1815 North University
Peoria, IL  61604
(309) 685-4011

John Hoffman
U.S. EPA
Office of Air and Radiation
401 M Street SW
Room 739 West Tower
Washington, D.C.  20460
(202) 382-4036

Don Johnson
Colorado State University
209 Animal Science Building
Ft. Collins, CO 80521
(303) 491-6672

Aslam  Khalil
Oregon Graduate Center
19600 N.W. Von Newman Dr.
Beverton, OR 97006-1999
(503) 690-1078

Ron Leng
Department of Biochemistry,
    Microbiology & Nutrition
The University of New England
Armidale, N.S.W. Australia  2351

Lisa Lewis
ICF Technology
10 Universal  City Plaza
Suite 2400
Universal City, CA 91608-1097
(818) 509-7150

Bob Lott
Gas Research Institute
8600 West Bryn Mawr Ave.
Chicago, IL  60631
(312) 399-8227
Richard Neal
National Cattlemen's Association
1301 Pennsylvania Ave., NW
Suite 300
Washington, D.C  20004
(202) 347-0228

John A. Patterson
Department of Animal Sciences
Purdue University
West Lafayette, IN  47907
(317) 494-4826

R.A. Rasmussen
Oregon Graduate Center
19600 N.W. Von Newman Dr.
Beverton, OR  97006-1999
(503) 690-1078

Joseph Robinson
The Upjohn Company
7922-190-MR
Kalamazoo, MI 49001
(616) 323-4000

Robert Roffler
Department of Animal Science
University of Idaho
Moscow, ID 83843
(208) 885-7173

Pat Russell
Department of Chemistry
University of California
Irvine,  CA  92717
(714) 856-5011

L.M. (Mac) Safley
Biological  and  Agricultural Engineering
North Carolina State University
Box 7625
Raleigh, NC 27695-7625
(919) 737-2694

-------
                           WORKSHOP ATTENDEES
                                     (Continued)
Lewis Smith
ARS-NPS
B-005 R 204
BARC-W
Beltsville, MD 20705
(301) 344-2737

Don Torell
American Sheep Industry Association
7950  Sanel  Dr.
Ukiah, CA  95482
(707) 462-9002
Henry Tyrrell
USDA Ruminant Nutrition Lab
BARC-East
Beltsville, MD  20705
(301) 344-2409

Patrick Zimmerman
NCAR
P.O. Box 3000
Boulder,  CO 80307
(303) 497-1000

-------