United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park, NC 27711
Research and Development
EPA/600/S7/91/003 Apr. 1991
i&EPA Project Summary
Global Warming Mitigation
Potential of Three Tree
Plantation Scenarios
R. L. Peer, D. L. Campbell, and W. G. Hohenstein
Increasing concentrations of carbon
dioxide (CO,) and other radiatively-im-
portant trace gases (RITGs) are of con-
cern due to their potential to alter the
Earth's climate. Some scientists, after
reviewing the results of general circula-
tion models, predict rising average
temperatures and alterations in the
Earth's hydrologic cycle. While the de-
bate continuesoverthe actual magnitude
of g lobal warming, most scientists agree
that some change will occur over the
next century. This places a burden on
policymakersto address global warming
and to develop mitigation measures. To
support the decision-making process,
the U.S. EPA's Air and Energy Engi-
neering Research Laboratory (AEERL)
is providing technical analyses of a va-
riety of global warming mitigation mea-
sures. This study analyzed alternative
uses of forests in the U.S. to reduce
atmospheric CO2 concentrations.
This Project Summary was developed
by EPA's Air and Energy Engineering
Research Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that is fully docu-
mented in a separate report of the same
title (see Project report ordering infor-
mation at back.)
Introduction
Since forests provide a sinkfor carbon by
fixing carbon dioxide (CO2) to produce bioj
mass, halting deforestation and creating
new forests have been proposed as means
of slowing the buildup of carbon'in the
atmosphere. However, using trees to scrub
CO2 from the atmosphere is a near-term
solution. During the early, high-growth phase
of life, a forest serves as a carbon sink. •
Eventually, the rate of growth slows, and the
death and decay of branches and leaves
begins to offset the carbon sink effect. F^
nally, as trees die and decompose, much of
the sequestered carbon returns to the at-
mosphere. An alternative is to harvest the
trees periodically and replant. This main-
tains the forest in its active growth phase,
maximizing the carbon uptake. In order for
this to be effective, the harvested wood
must1 be used in a way that conserves
RITGs. If the wood is used forf uel (replacing
fossil fuels) then, although CO2 is released,
no "new" CO2 is added to the atmosphere.
On the other hand, if it is used to make
disposable paper products, the carbon will
again be released into the atmosphere
without offsetting other CO, sources. If the
wood is used in a form that delays its eventual
decay and release to the atmosphere, then
some mitigative effect will be realized.
The purpose of this project was to ana-
lyze three reforestation scenarios that are
potential global warming mitigation methods:
(1) planting trees with no harvesting (NH),
(2) traditional forestry (TF), and (3) short-
rotation intensive culture (SRIC) of trees for
biomass. In addition to the cycling of CO2
through the trees, all other sources of CO2
and other RITGs associated with site
preparation, tree planting, harvesting, and
other activities specific to each scenario
also were estimated. The costs associated
with each scenario were estimated, and the
cost of using wood biomass as an alterna-
tive to fossil fuel was evaluated.
In this study, a common land base was
used to evaluate the three scenarios. In
r^x) Printed on Recycled Paper
-------
both the NH and TF scenarios, trees are
planted in plantations at densities that aver-
age 1,000 trees/ha. The SRIC scenario
assumes an average density of 2100 trees/
ha. The NH scenario assumes the tree
plantations are never harvested, but are left
to follow a natural successional pattern.
Trees are harvested every 6-8 years under
the SRIC scenario, compared to 35-80 year
rotations under the TF scenario. Existing
forest land was not included in the land
base. The land base included only crop and
pasture land in need of erosion control in the
U. S. A total of 40.4 million hectares in 10
geographfcalregions was used forthisstudy.
In the NH scenario, global warming is
mitigated by sequestering carbon in grow-
ing trees. In an actively growing forest,
carbon (as CO2) is removed from the atmo-
sphere at a much higher rate than it is
released (as CO2 or methane) by decom-
position. After some period of time, the
growth rate slows, dead biomass accu-
mulates, and decomposition processes
become more predominant. For this study,
it was assumed that a steady-state carbon
balance (i.e., no net flux) is reached at
maturity. The length of one rotation in tradi-
tional forestry was assumed to represent
the period of active growth. Therefore, in the
NH scenario, carbon is sequestered for a
period of time equal to the length of one TF
rotation for the region.
The TF scenario, in effect, extends the
carbon sink indefinitely by maintaining the
forest in the active growth phase. It is as-
sumed that the wood is used in such a way
that carbon is not immediately returned to
the atmosphere. Yields were derived from
published data and were assumed constant
over time. These same yields were used for
the NH scenario, but were assumed to
apply only to the young, rapidly growing
forest.
The Short Rotation Intensive Culture
(SRIC) scenario assumes that trees are
grown solely for the'production of biomass.
The biomass will be burned to produce
electricity, replacing coal as a fuel. In this
scenario, mitigation is achieved by the dis-
placement of coal emissions. Although
combustion of wood releases CO2, it isf ixed
in new plantations, resulting in no net in-
crease of CO2 in the atmosphere. If it is
assumed that coal would have been used to
produce the same amount of electricity,
then wood combustion actually results in
negative COZ emissions. Yields were esti-
mated for the next 20 years (near-term)
and, assuming continued research, for 20
years and beyond (mid-term).
In order to compare the SRIC and TF
scenarios better, the use of wood produced
under TF conditions as a fuel was also
analyzed. This is referred to as "TF burn."
Again, it is assumed that the wood would be
used in place of coal to produce electricity.
Air pollutants are emitted from forest
management activities due to machine use,
production and use of fertilizers and herbi-
cides, and the end-use of forest products.
Activities varied by scenario; for example,
harvesting occurred more often in the SRIC
scenario than in the TF scenario, and did
not occur at all in the NH scenario.
Table 1 lists the forest management ac-
tivities included in this analysis, and the
pollutants emitted from these activities that
were included in the analysis. A few emis-
sions were not included because the data
were inadequate to calculate a reliable
emission factor.
Emissions Analysis Results
The annual, emissions for each scenario
are shown in fable 2. The cumulative emis-
sions are also shown forthe years 2050 and
2100. The cumulativenumberswerederived
as follows:
• for SRIC, the near-term yields were as-
sumed for the first 20 years, the mid-term
yields thereafter;
• forTF and TF (burn), yields were assumed
constant overtime; and,
• for NH, TF yields were assumed through
2050, when carbon cycling was assumed
to reach a steady-state. VOCs continue
to be produced, however.
In Table 2, a negative number indicates a
sink, a positive number indicates a source.
Choosing the best mitigation scenario de-
pends on the criteria used. If CO2 reduction
alone is considered, the SRIC scenario is
clearly the most effective. This result is
driven entirely by the high yields assumed
for SRIC. Using TF-produced wood for
combustion is not nearly as effective, but
only because yields are lower.
The TF scenario does appear to be a
good long-term solution if only CO2 reduc-
tion is considered. However, the periodic
harvesting and planting emissions result in
greateremissionsof CO, CH4, N0]( NO, and
SO2 for the TF scenario than for the NH.
Since the first four are greenhouse gases
with radiative forcing values higher than
COg, the relative contribution of these
emissions should not be ignored. Further-
more, SO2 is a contributor to acid precipi-
tation. Overall, the NH scenario may be a
better choice for RITG reduction than the
TF.
The SRIC and TF (burn) scenarios result
in decreased CH4, NOx and SO2 emissions.
The last two are reduced because wood
combustion releases somewhat less NOX
and significantly less SO2 than coal com-
bustion. The CH4 reduction occurs because
less coal has to be mined (methane is
released when coal is mined).
All scenarios result in increased CO, VOC
and N2O. The increase in VQC comes al-
most entirely from the trees in the form of
terpenes and isoprenes. The increase in
CO is partly due to the combustion of diesel
fuel in the machinery used for planting and
harvesting, but is mostly attributableto wood
combustion. In the two cases where wood
replaces coal, a net increase, in CO occurs
because wood combustion produces rela-
tively high amounts of CO. Also, prescribed
burning in the TF scenario contributes some
CO. N2O is released due to the application
and degradation of nitrogenous fertilizers.
Cost Analysis Results
To adjust for differences in the rotation
length and annual yields between the in-
vestment scenarios, present net costs for
each investment scenario were found and
annualized over the investment's length.
The method used to annualize the invest-
ments converted cash streams, which were
variable over time, into even flow cash
streams. The annualized values were then
divided by the annual biomass yields to give
the annualized cost of producing one Mg of
biomass. These costs are reported in Table
3.
Management costs, including planting and
harvest costs, for traditional forestry and no
harvest scenarios were lower than for the
SRIC scenario. This is countered by higher
yields and shorter rotations for the SRIC
scenario. Biomass can be I grown more
cheaply underthe traditional forestry option
in all Southeast regions, the North Central
Lake States, and the Pacific Northwest.
Growing biomass using SRIC technologies
is competitive in the Pacific Northwest, the
Northeast, and North Central Non-lake
States. In the South Florida region, high
land costs also favor SRIC forestry (al-
though high land costs could lead to the
elimination of forestry altogether). Higher
annual expenditures in general tend to fa-
vor shorter rotations. •
Additional CO2 emissions savings can be
obtained by using biomass instead of fossil
fuels. Both electricity and ethanol can be
produced using wood as the feedstock.
These fuel costs are reported as a function
of feedstock price. Given the unit costs of
producing biomass under the scenarios,
the viability of producing electricity and
ethanol from wood was determined.
The costs of producing electricity from
wood biomass are reported in Table 4. In
order for biomass to be competitive with
coal for producing electricity, the biomass
must be available for less than $25.78/Mg.
This occurs only in the Pacific Northwest.
-------
Table 1. Forest Management Activities and Pollutants Emitted1
Activity
Planting
Fertilizer Production
Pesticide Production
Fertilizer Use
Hydrocarbons Emitted
from Trees
Prescribed Burning
Harvesting
Wood Transportation
Wood Combustion
Coal Mining (Displacement)
Coal Transportation
(Displacement)
Coal Combustion
(Displacement)
co,
X
X
X
X
X
X
X
X
X
CO
X
X
X
X
X
X
X
X
VOC
x
X
X
X
X
X
'
NO, CHt
X
X
X
x' x
X
X
X
X
X
X
so, NP
X
X
X
X
X
'Only those pollutants and activities quantified in this study are shown.
Table 2. Summary of Reforestation Scenarios: Emissions
SRIC
TF
TF(bum)
NH
-8.8E+07
-1.3E+07
-5400000
-1.6E+07
Total Annual Emissions (1000 Ma/Yr)
Scenario
SRIC
Near-term
Mid-term
TF
TF(burn)
NH
co,
-980000
-1700000
-210000
-90000
-260000
CO
2006.4
3045.7
2376.9
2597.8
0.2
voc
8037.9
8005.3
7884.5
7873.8
7740.1
CH4
-2867.0
-5038.4
104.0
-240.7
—
NO,
-1004.1
-1720.3
63.1
-81.2
1.5
N20
0.7
0.7
0.3
0.3
0.3
SO,
-4865.5
-8275.5
1.4
-566.9
1.0
176356
142614
155868
12
Total Emissions bv Year 2050 (1000 Ma)
480990
473070
472428
464406
-258876
6240
-14442
0
-8894
3786
-4872
90
42
18
18
18
-428330
84
-34014
60
Total Emissions bv Year 2100 (1000 Ma)
SRIC
TF
TF (bum)
NH
-1.7E+08
r2.3E+07
-990000
-1.6E+07
346641
271459
285758
12
881255
867295
866118
851411
-510796
11440
-26477
0
-174909
6941
-8932
90
77
33
33
18
-842105
154
-62359
60
-------
Table 3. Anthropogenic Emissions from Tree Plantation Scenarios Expressed as Percentage of 1985 NAPAP Anthropogenic
Emissions
Scenario
CO
NO
VOC
SRIC
Near-term 3.62
Mid-term 6.14
TF 4.29
TF(bum) 4.68
NH 0.00
1985 NAPAP Annual
Anthropogenic
Emissions 55,460
(1000 Mgfyear)'
-5.38
-9.21
0.34
-0.43
0.01
18,670
-23.21
-39.47
0.01
-2.70
OMO
20,960
-.23
-.39
.72
.67
0
20,084
Table 4. Cost of Electricity Production from Wood Biomass
(perMWh)
Region
South Florida
Southeast Coast
Southeast Piedmont
Southeast Mountains
Northeast '
North Central Lake States
North Central Non-Lake States
South Central Plains
Pacific Northwest-West
Pacific Northwest-East
Near-term
SRIC
$73.69
75.25
79.40
82.18
79.08
76.44
71.58
85.57
71.70
80.87
Mid-term
SRIC
$61.45
63.06
65.37
66.97
68.62
66.78
64.14
71.73
62.32
71.70
Traditional
Forestry
$73.71
57.78
57.14
58.65
84.45
59.53
72.88
' ..'•.
37.77
106.56
'Yields projected to be obtainable in 20 years.
However, as the technology of wood fired
power plants improves, the economics of
producing electricity from wood biomass
are likely to improve as well. If credits are
given to utilities for using wood instead of
coal, the economics could improve further.
Two methods of producing ethanol from
wood biomass were examined. The costs of
these methods were compared to those for
producing ethanolfrom corn. Forbothof the
wood based systems, the capital costs and
non-feedstock operating costs were too high
to make them competitive with ethanol pro-
duced from corn.
On a per acre basis, growing biomass
using traditional forestry methods appears
to be cheaper than SRIC methods. How-
ever, the total potential productivity of the
land is much higher for SRIC. Because of
this high productivity, SRIG appears to be
the best choice for mitigating emissions of
greenhouse gases. However, if a variety of
otherf actors are considered, the "best" miti-
gatiorv method is likely to be a composite
scenario with different methods imple-
mented in different regions.
-------
-------
R L Peer, D. L Campbell, and W. G. Hohenstein are with Radian Corp., Research
Triangle Park, NC 27709.
Christopher D. Geroni Is the EPA Project Officer (see below).
The complete report, entitled "Global Warming Mitigation Potential of Three Tree
Plantatbn Scenarios," (Order No. PB91-159 608/AS; Cost: $17.00, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati, OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA PERMIT NO. G-35
Official Business
Penalty for Private Use $300
EPA/600/S7/91/003
------- |