United States. Congress. House. Committee on Scien.

Road from Kyoto : hearing before the Committee on Science, U.S. House of Representatives, One Hundred Fifth Congress, second session (Volume pt. 2) online

. (page 21 of 137)
Online LibraryUnited States. Congress. House. Committee on ScienRoad from Kyoto : hearing before the Committee on Science, U.S. House of Representatives, One Hundred Fifth Congress, second session (Volume pt. 2) → online text (page 21 of 137)
Font size
QR-code for this ebook

at. (1991). and SA90. Emissions in the lEA and CEC
Scenarios are from 5% to 12% higher than the emissions in
the IS92a through to 2010. the end point of their scenarios.
Bumiaux ei at. (1991). using the OECD-GREEN model,
suggest CO2 emissions that are 4% lower than the IS92a in
the early years but increase to the same levels by 2020.
Emissions in the WEC moderate scenario are 24% lower
than lS92a in 2020. Figure A3.4 examines longer-term
scenarios, comparing the new IPCC Scenarios with 7
others, including the SA90 Scenario, two scenarios
(Slowly Changing World. SCW, and Rapidly Changing
World. RCW) developed by the US EPA (1990). scenarios
from Manne and Richels (1990). and several studies
underway within the Stanford University Energy Modeling
Forum. The modellers include Edmonds and Barnes. Peck
and Teisberg (using CETA). and Rutherford (using
CRTM) (Weyant. 1991: studies to be published in mid-
1992). The range of the new IPCC Scenarios is broader
than the "central tendency" studies presented in Figures

A3. 3 and A3.4. This is especially true by 2100. due to the
range of population and economic growth assumptions
used in the scenarios. Edmonds ei at. (1992) provides a
more detailed comparison of these scenarios.

A3.6 Energy

Future levels of greenhouse gas emissions from the energy
sector are a function primarily of population, incomes, the
structure and efficiency of economies, and the relative
costs and availability of different sources of energy. The
population and economic assumptions in this update of the
IPCC scenarios have already been discussed. To 2025. the
estimates of energy demand by region and sector are based
primarily on the EIS reference scenario. After 2025.
energy demand is a function of the economic assumptions
and the factors discussed in this Section, and modelled by
the ASF (EPA. 1990). Associated emissions of greenhouse
gases are estimated using coefficients from the OECD

The exogenous assumptions of improvements in the
intensity of energy end-use are critical parameters
counterweighing the upward push on CO^ emissions from
population and economic growth. The assumptions used
within IS92a and b result in a global decrease in energy
intensity of 0.8% annually in the period to 2025, and 1.0%
annually from 2025 to 2100. The decrease in energy per
unit of GNP is assumed to be particularly strong in China
through to 2100 and in Eastern Europe and the former
republics of the Soviet Union in the period 2000 to 2025 as
IS92a and b assume substantial structural change. This
reflects a complex mix of factors, including market-
oriented reforms, a tendency to increase energy demand
per capita with increased standards of living (though GNP
growth is substantially reduced in Eastern Europe and the


A3 Emissions Scenarios for the IPCC: an Update

use of non-fossil energy sources yield reductions in this
growth to an annual average of 0.2%. Emissions from
developing economies continue to rise due to increases in
population and economic growth. In the period. 1990 to
2100, CO2 emissions per capita from energy use grow at
an average annual rate of only 0.2% to 0.3% in the OECD,
Eastern Europe, and the republics of the former Soviet
Union, while averaging 1.0% for the rest of the world. CO2
emissions per capita in developing countries remain on
average one quarter to one half those of developed
countries by 2100. Conversely. CO2 emissions per dollar
GNP (per $GNP) in the OECD are two thirds the global
average, one half those of Eastern Europe and the former
Soviet Union, and one Fifth those of China. Global average
CO2 emissions per SGNP decline at an average annual rate
of 1.2% from 1990 to 2100. Regionally, the highest rate of
decline is in China where CO2 emissions per SGNP are
over four times higher than the global average in 1 990.

IS92b incorporates the stabilization goal for fossil
carbon dioxide emissions for the year 2000, proposed by
many OECD Member countries. If countries achieve these
commitments and sustain them through 2100 (which is
likely to require programmes beyond those already
planned) and the rest of the world does not adopt similar
measures, global emissions in 2025 of fossil carbon would
be 0.4 GtC lower than in the IS92a. This reduction
represents a reduction of 1 1% in emissions of CO2 from
the OECD from IS92a but only a 4% reduction of global
emissions. These results reflect the long-term contribution
of the economies of developing countries, the republics of
the former Soviet Union, and Eastern Europe to CO2

A3,7 Halocarbons

Halocarbons, including chlorofluorocarbons (CFCs). their
substitutes, and other compounds which deplete
stratospheric ozone, may have important implications for
climate change. Many of these compounds exert a much
more powerful direct radiative forcing than COi per
molecule. Recently, it has been discovered that the loss of
lower-stratospheric ozone can reduce the radiative forcing
of the troposphere/surface system, particularly at high
latitudes. Hence, ozone depleting molecules can have both
positive (direct) and negative (indirect) contributions to
radiative forcing. However, the net effect of such
halocarbons on globally averaged temperatures or, more
broadly, on climate is uncertain at present. As a result, the
comparisons of scenarios for these gases are summarized
using the kiloions (kt) of the compounds, not the index of
direct "Global Warming Potential" (GWP). as calculated
in Section A2 of this report.

An important event since the development of the
scenarios for the first IPCC assessment is the agreement to


adjust and amend the Montreal Protocol in London in 1990
(the "London Amendments"). Most key nations have
either now signed the agreement or have pronounced the
intention to do so. In IS92a, 70% of the developing world
is assumed to ratify and comply with the London
Amendments. This percentage is based on the GNP of
countries that have signed and/or ratified as of December
1, 1991 (e.g., China has signed while India has not). We
further assume in IS92a that if most of the world develops
and uses CFC substitutes, then the need to trade in global
markets and "technology transfer" will lead to a gradual
phase-out of all CFC use (we assume gradually from 2025
to 2075) even without worldwide ratification. We have
also included the voluntary reductions ahead of schedule
achieved by many countries. IS92b assumes global
compliance with the Montreal Protocol.

The London Amendments contain a recommendation
ooly to use halocarfoon substitutes for a transition period.
There are no international agreements for eventually
phasing down the production of all substitute compounds.
Some substitutes may not deplete stratospheric ozone but
may still contribute to climate change. Therefore, in the
scenarios, we assume that the production of substitutes
would mimic the growth rate of the underlying controlled
compounds which they replace under the phase-out.
adjusted for market reductions due to non-chemical
substitution and increased use of recycling and other
emission control programmes. Accordingly, all of the cases
assume that the demand for CFCs grows by 2.5% annually
until 2050 then remains flat. HCFCs and MFCs are
assumed to replace approximately 21 to 42% (depending
on the scenario) of phased-out CFCs. Substitution is
weighted much more towards HCFCs than HFCs over the
long-term unless additional policy steps are taken.

This analysis includes seven cases of future emissions of
CFCs and their substitutes. Three of these cases are
incorporated into the IS92 Scenarios. The first two cases.
"Partial Compliance and High HCFC" and "Partial
Compliance and Reduced HCFC" portray a future where
only 70% of the developing world ratifies and complies
with the London Amendments. In these scenarios, CFC
production In the remaining 30% continue to grow until
2100. Also, the HCFC reductions in the US are not
incorporated. The third scenario, "Partial Compliance and
Technology Transfer", is incorporated into IPCC Scenarios
IS92a, c, and f. It assumes partial compliance with the
London Amendments but assumes that "technology
transfer" results in a full phase-out of production of CFCs
by 2075. It includes the phase-out of HCFCs in the US
required by the Clean Air Act. The fourth case, "Global
Compliance", is incorporated in IS92b and contains full
global ratification and compliance with the London
Amendments. The fifth case, "97 Phase-Out for Developed
Countries", accelerates the phase-out schedule for CFCs.



Emissions Scenarios for the IPCC: an Update A3






iHgurc \3S:

Emission of chloronuorocarbons (CFCs) under a range of scenarios. Includes CFC- 1 1 . CFC- 1 2, CFC- 1 1 3. CFC- 1 14 and



2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

KEY :0CFC-11 »12 Q Omar CFCt ^ CO. []] MCF

III Halon 1211 & 1301 ^HCFCs Q HFO

Figure A3.6: Emission of halocarbons under IS92a. c and f. These assume only partial compliance with the London Amendments to
the Montreal Protocol but. through technical transfer to non-complying countries, a complete end to CFC production by AD 207S.

The sixth case. '•97 Phase-Out and HCFC Phase-Out" is
incorporated in IS92c and d and expands on the Tifth case
by incorporating a global phase-out of HCFCs. The
seventh case. "Faster Phase-Out - Global", accelerates the
phase-out of CFCs and HCFCs in developing countries.

For the calculations. US EPA's Integrated Assessment
Model for CFCs was used. Detailed results can be found in
the supporting document. Pepper ei al. (1992). In the
IS92a. emissions of CFCs. carbon tetrachloride, and
methyl chloroform decline rapidly through 2010 (see
Figures A3. 5 and A3. 6, "Partial Compliance and
Technology Transfer Case"). After 2010. these emissions

stabilize and ultimately decline to zero after 2075 as all the
world adopts the prominent technologies. Emissions of
HCFCs and HFCs grow rapidly throughout the whole time
horizon in all cases, reflecting their roles as substitutes for
the CFCs and the postulation of no controls on their
emissions (except on HCFCs in the US). Its results
indicate that the composition of substitutes could have an
important impact on levels of radiative forcing.

If "technology transfer" to non-signatories of the
agi^eement were not to lead to a phase-out of CFCs, the
implications by 2100 could be substantial, leading to
emissions almost back up to the level estimated for 2000.


A3 Emissions Scenarios for the tPCC: an Update

The results for the emissions of CFCs are depicted, as an
example, in the "Partial Compliance and High HCFC case
in Figure A3 J. In this case, after 2010, emissions of CFCs,
carbon tetrachloride and methyl chloroform stabilize and
then start to increase. This reflects growth from non-
signatories of the London Amendments.

IS92b assumes full ratification of and compliance with
the Lx>ndon Amendments. Emissions are nearly eliminated
much earlier ("London Amendments-Global Compliance")
than in IS92a. Moreover, recent data showing more
severe ozone depletion (WMO. 1992) could lead to a
more rapid phase-out of CFCs and halons, carbon
tetrachloride, and methyl chloroform ("97 Phase-Out for
Developed Countries"). Controls on the use of HCFCs are
also possible.

A3.8 Agriculture, Forests and Land Conversion
A3.8.1 Agriculture

Details of the estimation of greenhouse gases from
agricultural sources are available in Pepper ei al. (1992).
The distribution of the global emissions of non-CO^
greenhouse gases among different sources in the base year
has been taken from the assessment of the global budgets
reported in Section A I of this document. This distribution
is still poorly understood for most gases. In panicular,
emissions of CH4 from rice cultivation are highly
uncertain but lower than believed in IPCC (1990). Average
emission coefHcients of 38 grams per square metre per
year for land under rice cultivation have been selected for
all scenarios based on the CH4 emission budget of 60
teragrams (Tg). In lS92a and b. the emissions rise
gradually from 60Tg in 1990 to 88Tg by 2050. then
decline to 84Tg by 2 100. The scenarios assume continuing
advances in crop yields which average 0.5% annually over
the period. Consequently, the growth in emissions is
slower than growth in rice production, which more than


In IS92a and b, CH4 emissions from enteric fer-
mentation in domestic animals rise from 84Tg in 1990 to
close to 200Tg by 2100. This increase reflecu a rapid
increase in consumption of meat and dairy products and
assumes constant emissions per unit of production.
Emissions of CH4 from animal wastes have been added,
changing with the levels of meat and dairy production. If
meat production per animal were to increase, emissions
would be lower. Emissions from animal wastes increase
from 26Tg CH4 in 1990 to 62Tg CH4 by 2100. It is
uncertain whether this growth in the production of meat
and dairy products can actually be maintained, taking into
account possible land and feed constraints which are not
explicitly dealt with in these scenarios. Autonomous
developments that affect the emissions from enteric
fermentation or animal waste per unit of production, such
as those resulting from changed feeding panems, are not
hypothesized either. Both types of factors could change the
emission trends of these scenarios.

Emissions of N2O from fertilized soils in 1990 of 2.2
TgN have been selected as the starting budget, falling
within the range of uncertainty of 0.3 to 3.0 TgN reported
in Section A I . They increase in proportion to fertilizer use,
which more than doubles in IS92a and b. The impact of
changing fertilization practices and the dependency of N2O
emissions on local soil types, moisture, agricultural
practices, etc., has not been estimated.

A3.8.2 Forests and Land Conversion

Since SA90 was finalized, new data have become available
regarding both tropical deforestation rates and the average
content of carbon per hectare of above-ground vegetation.
Both are higher than the assumptions used in SA90. The
estimates of carbon in soils and fluxes of greenhouse gases
with changes in land uses remain as in SA90. Neither of
the possible effects on CO2 fluxes due to increased
fertilization or respiration, which may be associated with
higher CO2 atmospheric concentrations or temperature

Table A3.8: Assumptions used in deforestation cases.

Scenario Used In

Biomass Content

Rate of Deforesution

IS92a. b. & e
















Moderate (tied to moderate population growth)

Moderate (tied to low population growth)

Halt Tropical Deforestation

Moderate (tied to high population growth)

Moderaie (tied to moderaie population growth)



Halt Tropical Deforestation/Increase Establishment of Plantations



Emissions Scenarios for the IPCC: an Update A3

increases, have been incorporated in this analysis.

To incorporate the new data and the uncertainties still
surrounding these parameters, eight cases of tropical forest
clearing and emissions of greenhouse gases were
developed. These eight cases include four cases which
were incorporated within the new IPCC Scenarios which
assume moderate assumptions of rates of tropical
deforestation (except IS92d which has a halt to defor-
estation) and biomass content of vegetation in these
forests. The eight cases also include four sensitivities
around the case incorporated in IS92a. The sensitivity
cases vary rates of deforestation, rates of establishment of
plantations, and assumptions concerning the biomass
content of the forests. Table A3. 8 summarizes these cases
and their assumptions.

As lands convert from one use to another, greenhouse
gases can be released or taken up by vegetation and soils,
for example by the burning or regrowth of forests or the
tilling or amendment of soils. This analysis simulates and
tracks the changes of land parcels from one use to another
from 1975 to 2100, due to agricultural demand, burning,
plantations, etc., and calculates the associated greenhouse
gas emissions and uptake over time. As land is cleared,
sometimes more than once in the period of analysis, only
pan of the carbon stored in vegetation and soils is released
over an extended period of time. As regrowth occurs on
cleared land, carbon is sequestered. We calculate the net
balance of carbon from vegetation and soils of all lands
estimated to be tropical forests at any time from 1975
through 2100. Assumptions about rates of carbon loss or
absorption and other parameters used in this analysis, as
well as the case results, are detailed in Pepper et al. ( 1992).

The 1S92 Scenarios use the new FAO 1990 Tropical
Forest Assessment (FAO, 1991) and the 1988 update of
the 1980 Tropical Forest Assessment (FAO. 1988) for its
deforestation rates. The new assessment estimates that, on
average. 17 million hectares of tropical closed and open
forest were cleared annually from 1981 to 1990. The 1988
FAO Tropical Forest Assessment provided estimates of
clearing rates for the period 1976 to 1980. While questions
have been raised concerning the reliability of these data,
they are the best and most recent currently available for the

We estimated a constant rate of change in clearing rates
over this period such that the average quantities of clearing
for 1976 to 1980 and for 1981 to 1990 match those
reported in the FAO 1988 and 1991 reports, respectively.
Moreover, this estimate is constrained so that forest areas
in 1980 equal the quantities given in the more complete
and detailed 1980 assessment. This results in calculated
clearing rales increasing from 13.2 million hectares in
1980 to 19.3 million hectares in 1990. After 1990.
deforestation rates increase in proportion to population, but
lagged twenty years and constrained by available forest

area in each country. In the "high deforestation" sensitivity
cases, these rates are increased by an additional 1% point
per year. In the "halt deforestation" sensitivity case and in
IS92d. we assume that rates of deforestation decline,
starting in 1990. The IPCC Greenhouse Gas Task Force
advised that it should be assumed that all forests not
legally protected, including areas which have been
classiTied as non-productive, can be subject to defor-
estation (IPCC. 1991c).

Forest clearing in lS92a, b, c, e, and f increases to 20 to
23.6 million hectares per year by 2025. depending on
population growth, and then declines. In lS92d. clearing
declines steadily after 1990 to 0.7 million hectares per year
by 2025. In the high deforestation sensitivity cases,
tropical forest clearing increases to 28.6 million hectares in
2025 before declining. In all sets of assumptions, available
forest resources within each country provide upper bounds
on future clearing. In the IS92a. 73% of all tropical forests
(1.4 billion hectares), are cleared by 2100. In the high
deforestation sensitivity case, this fraction increases to
91%. In IS92a, countries representing 43% of forest
clearing in 1980 have (or have nearly) exhausted their
forest resources by 2025. By 2050, this fraction increases
to 52%.

Other factors which vary in the sensitivity cases include
the fate of forest fallow, future rates of plantation
establishment, and the carbon stored in the aboveground
biomass. The high deforestation sensitivity cases include
another possible net source of carbon; permanent clearing
of forest fallow. These are areas of logged or abandoned
agricultural lands which are regenerating to forest. The
high deforestation sensitivity cases assume that up. to 10
million hectares of forest fallow are currently being
convened to permanent agriculture annually (Houghton.
1991) and that this clearing continues into the future until
almost all forest fallow is converted. In IS92a, the
establishment of plantations, which FAO (1988) estimates
as 1.3 million hectares annually between 1980 and 1985, is
assumed to continue with 1 18 million hectares added
between 1990 and 2100. The high deforestation sensitivity
cases assume that no new plantations are added after 1990.

Moderate estimates of carbon stored in the biomass are
from OECD (1991) which have been adapted from Brown
and Lugo (1984), Brown el al. (1989), and Brown (1991),
and have been estimated using wood volumes. The
Moderate biomass estimates are used for the IS92
Scenarios. The high biomass estimates, used in the
sensitivity analyses, increase the moderate estimates by the
percentage corrections cited in Houghton (1991) to reflect
uncertainties in measureinenl techniques and results from
other studies utilizing alternative (i.e.. destructive
sampling) approaches. Table A3.9 summarizes current
estimates of biomass contents and those used in all IS92


A3 Emissions Scenarios for the IPCC: an Update
Table A3.9: Carbon stored in tropical forests (tons Clhectare).



.... IS92 a,b ^^^

Closed Open
Conif. B-leaf


Carbon Stocks from Houghton (1991)
Earlier Estimate ^ Recent Estimate ^^

Moist Seasonal Closed Open

Forest Forest Forest Forest Crops

(D/V)t (D/V)t (U/L)^' (DA')t


Latin America 76

Asia 97

Africa 117

Soils 100

































* For coluinns labelled (D/V), the first value is based on destructive sampling of biomass and the second value is

calculated from estimates of wood volumes.
' ' For columns labelled (U/L). the first value is for undisturbed forests and the second value is for logged forests,
ttt Source:OECD(l99l).


Figure A3.7: Comparison of net CO^ emissions from tropical deforesution according to some longer-range scenarios: ( I ) High
deforestation/high biomass: (2) High deforestaiion/moderaie biomass: (3) moderate deforesiaiion/high biomass: (4) IS92f: (5) IS92a. b
and e - moderate deroresuiion/modeiaie biomass: (6) IS92c: (7) lS92d. and (8) No deroresiation/high plantation. (Note: does not
include the effects of fertilization or increased respiration on net COj emissions due to higher COy atmospheric concentrations.)

CO2 emissions in the IS92 Scenarios and (he sensitivity
cases range from I.I GtC to 2.3 GtC in 1990. These
Include net soil carbon released as well. In the IS92a.
emissions are relatively flat through to 199S and then start
to decline. Net emissions are slightly negative by 2 100 due
to carbon sequestration by plantations. In the high
deforestation and high biomass sensitivity case, emissions
increase to 3.0 GtC by 2025. decline to 1.9 GtC in 2050
and 0.2 GtC by 2100. Even though in all of the cases
clearing of open forests in the period 1980 to 1990
represents over one third of total clearing, net emissions of
COi from open forests represent less than 10% of total
deforestation emissions.

All these cases do not span the range of possibilities for
both current and future emissions. There is a strong need

for improvement of the base data. Moreover, current trends
in deforested area, combined with changes in policies,
could very well lead to emissions lower than in (he
moderate case.

In conclusion, the sensitivity analysis explores a wide
range of possible futures and identines the importance of
several key assumptions. Figure A3.7 illustrates the CO2
emissions from the IS92 Scenarios along with the
sensitivities around lS92a. One sensitivity case, "High
Deforestation/Moderate Biomass" illustrates the impact of
the higher clearing rates and forest fallow clearing
assumptions on COi emissions. The sensitivity case,
"Moderate Deforestation/High Biomass", illustrates the
importance of assumptions concerning carbon stored
within the biomass. The sensitivity case. "High Defor-



esiation/High Biomass" shows the combined impact of
these alternative assumptions. The "Halt Deforestation/
High Plantations" sensitivity case illustrates the potential
for reducing emissions by quickly stopping forest clearing
and actively establishing plantations. In this case, forest
clearing is reduced starting in 1991 and eliminated by
2025. Up to 293 million hectares of plantations are

Online LibraryUnited States. Congress. House. Committee on ScienRoad from Kyoto : hearing before the Committee on Science, U.S. House of Representatives, One Hundred Fifth Congress, second session (Volume pt. 2) → online text (page 21 of 137)