rw^ ;
ff
S;i
S£
United States Environmental Office of Atmospheric Programs (6207J) EPA 430-R-05-006
Protection Agency Washington, DC 20460 November 2005
Greenhouse Gas Mitigation
Potential in U.S. Forestry
and Agriculture
'$&?.
-*t-,
^i«v
«^(^i!
iM
~''>jr.
"^!|^
^T^'-
??
-------
,' ".,,IHS«v5*
P''
I
BE
I
1
How to obtain copies
You can electronically download this
document from U.S. EPA's web page at
http://www.epa.gov/sequestration. To request
free copies of this report, call the National Service
Center for Environmental Publications (NSCEP)
at 1 - (800) 490-9198.
For further information
For further information, contact Kenneth
Andrasko, (202) 343-9281, andrasko.ken@epa.gov,
or Benjamin DeAngelo, (202) 343-9107,
deangelo.ben@epa.gov, U.S. Environmental
Protection Agency.
-------
Greenhouse Gas Mitigation Potential
in U.S. Forestry and Agriculture
November 2005
United States Environmental Protection Agency
Office of Atmospheric Programs (6207J)
1200 Pennsylvania Ave., NW
Washington, DC 20460
-------
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
Acknowledgments
1GG&
|5j
This report was prepared under a contract between the U.S.
Environmental Protection Agency (EPA) and RTI International.
The main authors of the report are Brian C. Murray, Brent
Sohngen,1 Allan J. Sommer, Brooks Depro, and Kelly Jones
of RTI; Bruce McCarl of Texas A&M University and Dhazn
Gillig of American Express Corporation;2 and Benjamin
DeAngelo and Kenneth Andrasko of EPA. The report was
edited by Kenneth Andrasko and Benjamin DeAngelo of EPA.
The authors acknowledge the FASOMGHG model develop-
ment efforts over the past decade of Darius Adams of Oregon
State University; Ralph J. Alig of the USDA Forest Service in
Corvallis, OR; John "Mac" Galloway, UNEP Risoe Centre on
Energy, Climate and Sustainable Development; and Steven
Winnett, EPA.
We thank the following external reviewers: Richard Birdsey,
USDA Forest Service; John Brenner, USDA Natural Resources
Conservation Service; Suzie Greenhalgh, World Resources
Institute; Cesar Izaurralde, Pacific Northwest National
Laboratory; Jan Lewandrowski, USDA Office of the Chief
Economist; Ruben Lubowski, USDA Economic Research
Service; Michelle Manion, Union of Concerned Scientists;
Reid Miner, National Council for Air and Stream Improve-
ment; Sian Mooney, University of Wyoming; Keith Paustian,
Colorado State University; Neil Sampson, The Sampson
Group; Ron Sands, Pacific Northwest National Laboratory;
and Tristram West, Oak Ridge National Laboratory. We also
thank other EPA reviewers: Steven Rose, Francisco de la
Chesnaye, Dina Kruger, Allen Fawcett, and John Powers.
Research assistance was provided by Laurel Clayton and
Catherine Corey of RTI. Sharon Barrell of RTI coordinated
editing and publications support.
1 Dr. Sohngen was on sabbatical from The Ohio State University when
working on this report at RTI.
2 Dr. Gillig was at Texas A&M University when she performed this work
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
Table of Contents
Acknowledgments i
Table of Contents iii
List of Tables, Figures, and Boxes vii
Tables vii
Figures viii
Boxes x
Executive Summary ES-1
1. Introduction 1-1
Purpose and Approach of this Report 1-3
Organization of Report 1-3
2. Greenhouse Gas Mitigation Options in U.S. Forestry and Agriculture 2-1
Chapter 2 Summary 2-1
Carbon Sequestration 2-1
Afforestation 2-2
Forest Management 2-2
Agricultural Soil Carbon Sequestration 2-5
Grassland Conversion 2-5
Grazing Management 2-5
Riparian Buffers 2-6
GHG Emissions Reduction Options in Agriculture 2-6
Reduction of CO2 Emissions from Fossil Fuel Use 2-6
Reduction of Non-CO2 GHG Emissions 2-6
Biofuel Offsets of Fossil Fuels 2-9
Unique Time Dynamics of Carbon Sequestration Options 2-9
"Saturation" of Carbon Sequestration to Equilibrium 2-9
Reversibility of Carbon Sequestration 2-11
Accounting for Carbon after Timber Harvests 2-12
Addressing Carbon Sequestration Dynamics in this Report 2-12
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE iii
-------
3. Modeling Framework and Baseline 3-1
Chapter 3 Summary 3-1
Modeling Framework 3-1
General Model Description 3-2
Geographic Coverage/Regional Detail 3-4
Land Base 3-4
General Economic Concepts: Optimizing Behavior 3-4
Forest-Sector Economic Detail 3-6
Agriculture-Sector Economic Detail 3-8
Biofuels 3-9
Cross-Sector Land Interaction 3-10
Greenhouse Gas Accounting 3-10
Non-GHG Environmental Indicators 3-13
GHG Mitigation Strategies 3-13
Baseline GHG Projections from the Forest and Agriculture Sectors 3-15
FASOMGHG Baseline Projections 3-15
Comparison of FASOMGHG Baseline GHG Projection to Other Published Estimates .. 3-19
Applying FASOMGHG for the Purposes of this Report 3-24
4. Mitigation Potential: Comprehensive Scenarios with All Activities and All GHGs 4-1
Chapter 4 Summary 4-1
Mitigation Responses under Various GHG Mitigation Scenarios 4-2
Scenarios Description: Constant and Rising Incentives for GHG Mitigation 4-2
Mitigation Response to Constant GHG Price Scenarios 4-5
Mitigation Response to Rising GHG Price Scenarios 4-18
Comparison of FASOMGHG Results with Other Analyses 4-21
Richards and Stokes (2004): Forest Carbon 4-21
Stavins (1999): Afforestation 4-22
Sedjo, Sohngen, and Mendelsohn (2001): Forest Carbon 4-23
USDA, Economic Research Service (2004): Agricultural Carbon Sequestration 4-24
Recap of Study Comparisons 4-24
Appendix 4.A 4-25
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
5. Mitigation Potential of Selected Activities 5-1
Chapter 5 Summary 5-1
Fixed Quantities of National GHG Mitigation 5-2
National-Level Results by Activity and Time Period 5-2
Regional Activity Contributions to National Mitigation Levels 5-7
National Mitigation Quantity Scenarios Summary 5-8
Limiting Payments by GHG Type 5-8
Paying for CO2 Only vs. Paying for All GHGs: $15/t CO2 Eq 5-8
CO2 Only: Mitigation Over Time 5-9
Selected Activity Scenarios 5-9
National Results 5-11
Regional Results 5-11
6. Implications of Mitigation via Selected Activities 6-1
Chapter 6 Summary 6-1
Project Quantification Issues and Costs 6-2
Quantifying the Net GHG Contribution of Projects 6-2
Other Project Implementation Considerations 6-11
Preliminary Assessment of Implementation Factors by Major Mitigation Activity 6-13
Per-Acre Payments for Carbon Sequestration to Address Measurement Difficulties 6-14
Scenario Description 6-14
Per-Acre Payments for Carbon Sequestered through Afforestation 6-16
Per-Acre Payments for Agricultural Soil Carbon Sequestered through
Changes in Tillage 6-17
7. Non-GHG Environmental Co-effects of Mitigation 7-1
Chapter 7 Summary 7-1
Land Use 7-1
Regional Distribution of Land Uses 7-2
Timberland Management Intensity 7-5
Agricultural Nonpoint Pollutant Runoff 7-5
Changes in Agricultural Runoff and Water Quality—Results from a Separate
Case Study 7-8
Implications for Biodiversity of GHG Mitigation 7-11
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE v
-------
8. Summary of Insights on Key GHG Mitigation Issues 8-1
Key Issues 8-1
Insights from Analyzed Results 8-2
While national mitigation rates decline over time (under constant price scenarios),
cumulative GHG mitigation steadily increases 8-2
Identifying attractive activities may require looking at a range of characteristics
for each option 8-3
The quantity and timing of mitigation can determine the selected activities 8-3
Achieving a specific mitigation level within a narrow time frame may shift
emissions to periods before and after the period of interest 8-3
Under scenarios of rising GHG payments, forest and agriculture mitigation action
may be delayed 8-6
GHG incentives reduce net emissions from the forest and agriculture sectors below
baseline levels. If the incentives are strong enough, the joint sectors could move
from a net emissions source to a sink 8-6
Leakage potential from limiting included mitigation activities may be largely
confined to the forest sector 8-7
Raising GHG mitigation levels in forestry and agriculture can cause environmental
co-effects, both good and bad 8-8
Payment method will determine efficiency of mitigation activities 8-8
If outreach is needed to deliver GHG mitigation, these efforts might focus in regions
with the largest mitigation potential 8-9
9. References.. . R-1
vi GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
List of Tables, Figures, and Boxes
Tables
Table 2-1: Representative Carbon Sequestration Rates and Saturation Periods
for Key Agriculture, Land-Use Change, and Forestry Practices 2-3
Table 2-2: Agricultural Non-CO2 Emissions by Source, 2003 (Tg CO2 Eq.) 2-7
Table 3-1: FASOMGHG Model: Key Dimensions 3-3
Table 3-2: FASOMGHG Regional Definitions 3-5
Table 3-3: Agriculture-Sector Commodities 3-8
Table 3-4: GHG Emission Sources and Sinks in FASOMGHG 3-11
Table 3-5: Broad GHG Mitigation Strategies Covered in FASOMGHG 3-13
Table 3-6: Mitigation Options Not Explicitly Captured in FASOMGHG 3-14
Table 3-7: U.S. Land-Use Change for Major Categories: 1982-1997 3-16
Table 3-8: Baseline Forest and Agriculture GHG Net Annual Emissions by Activity
and Decade for the United States: FASOMGHG Model: 2010-2050 3-18
Table 3-9: Net Annual CO2 Flux from U.S. Forest Carbon Stocks: 1990 and 2000,
EPA Inventory Quantities (in Tg CO2 per year) 3-20
Table 3-10: Projected Net CO2 Flux from U.S. Forest Carbon Stocks: 1990-2040,
USDA Forest Service Estimate 3-20
Table 3-11: Non-CO2 GHG Emissions from Agriculture (Tg CO2 Eq.): EPA GHG Inventory,
1990-2003 3-23
Table 4-1: Core Price Scenarios 4-3
Table 4-2: CO2 and C Price Equivalents 4-3
Table 4-3: Acreage Converted from Conventional Tillage to Reduced Tillage under Baseline
and GHG Prices: U.S. Total (Million acres) 4-7
Table 4-4: Comparison of Annualized GHG Mitigation Estimates (Tg CO2 Eq. per year)
across Alternative Time Horizons at a GHG Price of $15/t CO2 Eq 4-11
Table 4-5: National GHG Mitigation Totals by Activity: Annualized Averages, 2010-2110 4-12
Table 4-6: Top 10 Region-Activity Mitigation Combinations 4-17
Table 4-7: Comparison of FASOMGHG Results in this Chapter to Range of Estimates from
Richards and Stokes' (2004) Review Study 4-21
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE vii
-------
Table 4-8: Comparison of FASOMGHG Results in this Chapter to Stavins' (1999) Study 4-22
Table 4-9: Comparison of FASOMGHG Forest Carbon Sequestration Results in this
Chapter with Sedjo, Sohngen, and Mendelsohn (2001) 4-23
Table 4-10: Comparison of this Study with Lewandrowski et al. (2004) (USDA ERS) 4-24
Table 4.A.I: Key Results at the National Level by Activity, Time Period, and Constant-Price
Scenarios 4-25
Table 4.A.2: Total Forest and Agricultural GHG Mitigation by Region 4-26
Table 4.A.3: Forest and Agricultural GHG Mitigation by Activity, Region, and Price Scenario ... 4-26
Table 4.A.4: Key Results at the National Level by Activity, Time Period, and Rising Price
Scenarios 4-28
Table 5-1: National GHG Mitigation Quantity Scenarios for 2025 and 2055 5-2
Table 5-2: National Mitigation, by Scenario and Activity, for Least-Cost Quantity in 2025
and 2055: Annualized over 2010-2110 5-3
Table 5-3: Least-Cost Mitigation Response to Fixed National GHG Mitigation Levels in
2015, 2025, and 2055 5-6
Table 5-4: GHG Mitigation Quantity Ranking by Region/Activity Combination: Fixed
National Mitigation Quantity Scenarios 5-7
Table 5-5: Mitigation Quantities: Payments for CO2 Only vs. Payment for All GHGs
($15 per t CO2 Eq.) 5-8
Table 5-6: National GHG Mitigation Totals in Key Years by Activity: Payment for CO2
Only at $15/t CO2 Eq. (Includes Non-CO2 GHGs) 5-9
Table 5-7: Selected Activity Scenarios 5-10
Table 5-8: GHG Mitigation under Payment for Specific Activity Scenarios 5-11
Table 6-1: Candidate Approaches for Accounting for Reversal Risk from Carbon-Based
GHG Mitigation Projects 6-4
Table 6-2: Leakage Estimates by Mitigation Activity at a GHG Price of $15/t CO2 Eq 6-6
Table 6-3: Afforestation Regional Leakage Estimates from Murray et al. (2004) 6-9
Table 6-4: Forest Preservation and Avoided Deforestation Regional Leakage Results
from Murray et al. (2004) 6-9
Table 6-5: Implementation Issues for Selected Activities and Projects: Leakage Estimates
from FASOMGHG and MMV . . 6-14
viii
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
Table 6-6: Qualitative Consideration of Implementation Issues for Selected Activities
and Projects: Baselines, Additionality, and Reversal Risk 6-15
Table 6-7: Per-Acre vs. Per-Tonne Payment Approaches for Afforestation: 2015 and
2010-2110 Annualized 6-16
Table 6-8: Agricultural Soil Carbon Sequestration Payment Approaches: 2015 and
2010-2110 Annualized 6-18
Table 7-1: Land Use under the Baseline, $15, and $50 (Constant) GHG Price Scenarios:
2015 and 2055 7-2
Table 7-2: Change in Pollutant Loadings for Selected Agricultural Pollutants and the
WQI for the $6.80 per tonne CO2 Eq. Scenario, using the ASMGHG-NWPCAM
Model Integration 7-10
Table 8-1: Characteristics of GHG Mitigation Activities 8-4
Table 8-2: Potential Implications of Mitigation Level and Time Frame 8-4
Table 8-3: Leakage Estimates by Mitigation Activity at a GHG Price of $15/t CO2Eq 8-7
Figures
Figure 1-1: Forestry and Agriculture Net Contribution to GHG Emissions in the United
States, 2003 1-2
Figure 2-1: Agricultural Non-CO2 Emissions by Source Relative to All Other GHG Emissions... 2-7
Figure 2-2: Conceptual Model of Soil Organic Matter Decomposition and Accumulation
Following Disturbance 2-10
Figure 2-3: Absolute Change in the Annual Rate of Carbon Sequestered Following a
Change from Conventional Tillage (CT) to No-Till (NT) 2-11
Figure 2-4: Carbon Accumulation on an Afforested Stand to Saturation 2-11
Figure 2-5: Cumulative Carbon Changes for a Scenario Involving Afforestation and Harvest... 2-12
Figure 3-1: FASOMGHG Regions 3-5
Figure 3-2: FASOMGHG Market Linkages 3-10
Figure 3-3: Cumulative Carbon Changes for a Scenario Involving Afforestation and Harvest... 3-12
Figure 3-4: Baseline Land-Use Projections, FASOMGHG: 2010-2050 (Million acres) 3-15
Figure 3-5: Total Factor Productivity in U.S. Agriculture: 1949-1998 3-17
Figure 3-6: Forest and Agriculture Products Price Series 3-17
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
Figure 3-7: Comparison of Projected Baseline Carbon Sequestration Trends in U.S. Forests:
FASOMGHG vs. USDA Forest Service Model 3-21
Figure 3-8: Comparison of Projected Baseline Non-CO2 GHG: FASOMGHG vs. Scheehle
and Kruger (in press) 3-23
Figure 4-1: Price Trajectories for Rising-Price Scenarios 4-4
Figure 4-2: Land Use in 2025 at Different GHG Price Levels 4-6
Figure 4-3: Timberland Area over Time: $50/t CO2 Eq. vs. Baseline 4-6
Figure 4-4: Effect of GHG Prices on Forest Management Variables, 2015 4-8
Figure 4-5: National GHG Mitigation at Representative Years by Price (2015, 2025, and 2055) 4-8
Figure 4-6: Cumulative GHG Mitigation over Time 4-9
Figure 4-7: Comparison of Actual, Cumulative Average, and Annualized GHG Mitigation
Value Calculations at $15/t CO2 Eq.: 2010-2110 4-11
Figure 4-8: GHG Mitigation Supply Function from National GHG Mitigation Totals
by Activity 4-12
Figure 4-9: Model Sensitivity to Saturation Period toward a New Soil Carbon Equilibrium
from Tillage Change: GHG Price = $15/t CO2 Eq 4-15
Figure 4-10: Sensitivity of Model Results to Assumed Biofuel Demand Restrictions: GHG
Price = $30/t CO2 Eq 4-15
Figure 4-11: Total Forest and Agriculture GHG Mitigation by Region 4-16
Figure 4-12: Pollutant Loading Effects Over Time of a $15/t CO2 Eq. GHG Price 4-18
Figure 4-13: Constant-Price Scenarios vs. Rising-Price Scenarios and GHG Mitigation 4-19
Figure 4-14: Cumulative GHG Mitigation over Time: $3/t CO2 Price Rising at Two Rates 4-20
Figure 4-15: Cumulative GHG Mitigation over Time: $20/t CO2 Price Rising by $1.30 per
Year ($75 cap) 4-20
Figure 5-1: Least-Cost Mitigation Quantities by Scenario and Activity in 2025 and 2055 5-3
Figure 5-2: Scenarios with Objective of Mitigating: (a) 375 Tg CO2 Eq. in 2025 and
Maintaining; (b) 375 in 2025 and 900 Tg CO2 Eq. in 2055; and (c) 375 Tg CO2
Eq. in 2025 without Maintaining Thereafter 5-5
Figure 5-3: Cumulative Mitigation: Payment for CO2 Only (Includes Non-CO2 GHGs)
vs. All GHGs at $15/t CO2 Eq 5-10
Figure 5-4: GHG Mitigation under Payments for Afforestation and Forest Management
Only at $15/t CO2 Eq.: By Region 5-12
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
Figure 5-5: GHG Mitigation under Payments for Biofuel Offsets Only at $3/t CO2 Eq.,
Rising at 4 Percent per Year, By Region 5-13
Figure 5-6: GHG Mitigation by Region and Activity under Payments for Agricultural
Management Only: $15/t CO2 Eq 5-13
Figure 5-7: Regional Distribution of Soil Carbon Sequestration under Payment for Soil
Carbon Only: $15/t CO2 Eq. Constant Price 5-14
Figure 6-1: Regional Leakage Flows for Afforestation-Only Payment Scenario: $15/t CO2 Eq 6-8
Figure 6-2: Regional Shares of Afforestation Carbon Sequestration by Payment Approach 6-17
Figure 6-3: Regional Shares of Agricultural Soil Carbon Sequestration by Payment Approach .. 6-18
Figure 7-la: Land-Use Allocation by Eastern U.S. Regions in 2015: Baseline and the $15 and
$50 Constant GHG Price Scenarios 7-3
Figure 7-lb: Land-Use Allocation by Eastern U.S. Regions in 2055: Baseline and the $15 and
$50 Constant GHG Price Scenarios 7-3
Figure 7-2a: Land-Use Allocation by Western U.S. Regions in 2015: Baseline and the $15 and
$50 Constant GHG Price Scenarios 7-4
Figure 7-2b: Land-Use Allocation by Western U.S. Regions in 2055: Baseline and the $15 and
$50 Constant GHG Price Scenarios 7-4
Figure 7-3: Soil Erosion Index over Time by (Constant) GHG Price Scenario (Baseline = 100).... 7-6
Figure 7-4: Phosphorous Loading Index over Time by (Constant) GHG Price Scenario
(Baseline = 100) 7-7
Figure 7-5: Nitrogen Runoff Index over Time by (Constant) GHG Price Scenario
(Baseline = 100) 7-8
Figure 7-6: Pesticide Index over Time by (Constant) GHG Price Scenario (Baseline = 100) 7-8
Figure 7-7: Changes in Water Quality from Soil Carbon Sequestration and Other Agricultural
Management Changes under $6.8 per Tonne CO2 Scenario in ca. 2020, using the
ASMGHG-NWPCAM Integrated Agriculture Water Quality Model 7-10
Figure 8-1: National GHG Mitigation at Three Focus Dates by GHG Price: Average Annual 8-2
Figure 8-2: Cumulative GHG Mitigation in Tg CO2 Eq 8-3
Figure 8-3: Responses to Set Mitigation Quantities: Cumulative Mitigation to 2100 8-5
Figure 8-4: Constant Price Scenarios vs. Rising Price Scenarios and GHG Mitigation 8-6
Figure 8-5: Cumulative Net Emissions/Sinks for Forestry and Agriculture: Comparison of
Baseline and Comprehensive Mitigation Scenarios at Constant Prices over Time .... 8-7
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
Figure 8-6: Nitrogen Runoff Index over Time by (Constant) GHG Price Scenario 8-8
Figure 8-7: Total Forest and Agriculture GHG Mitigation by Region 8-9
Boxes
Box 1-1: Relative Global Warming Potential of Non-CO2 Gases 1-1
Box 3-1: Perfect Foresight in Climate Economic Models 3-6
Box 4-1: Measurement Units Reported in the Analysis 4-2
Box 4-2: Methods Used for Reporting GHG Mitigation Results at Different Points in Time.... 4-3
Box 4-3: Technical, Economic, and Competitive Potential of a GHG Mitigation Option 4-5
Box 4-4: Summary of Constant GHG Price Scenario Results 4-5
Box 4-5: Annualizing Results over the Projection Period 4-10
Box 4-6: Sensitivity Analysis of Key Assumption: Time to Reach Soil Carbon
Equilibrium ("Saturation") 4-14
Box 4-7: Sensitivity Analysis of Key Assumption: Biofuel Demand 4-15
Box 6-1: Shortening the Time Horizon for Quantifying Leakage 6-7
xii GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
Executive Summary
Forestry and agricultural activities are widely
recognized as potential greenhouse gas
(GHG) mitigation options. Activities in
forestry and agriculture can reduce and avoid the
atmospheric buildup of the three most prevalent
GHGs directly emitted by human actions: carbon
dioxide (CO2), methane (CH4), and nitrous oxide
(N2O). The removal of atmospheric CO2 through
sequestration in carbon "sinks" is a mitigation
option in forestry and agriculture that has received
particular attention.
Currently in the United States, forest and agricul-
tural lands comprise a net carbon sink of almost
830 teragrams (Tg or million tonnes1) of CO2
equivalent (or nearly 225 Tg of carbon equivalent)
per year, according to the U.S. GHG inventory
(EPA 2005). Removal of atmospheric CO2 through
carbon sequestration is greater than CO2 emissions
from events such as forest harvests, land conver-
sion to other uses, or fire. The U.S. net carbon
sink—over 90 percent of which occurs on forest
lands—currently offsets 12 percent of U.S. GHG
emissions from all sectors of the economy on an
annual basis (EPA 2005). The agriculture sector,
however, is a net emitter of GHGs. Agricultural
CH4 and N2O emissions are responsible for over
6 percent of all annual U.S. GHG emissions (EPA
2005). After accounting for both carbon sequestra-
tion and non-CO2 emissions, the forest and agricul-
ture sectors comprise a net GHG sink that offsets
almost 6 percent of total U.S. GHG emissions.
This report evaluates the potential for additional
carbon sequestration and GHG reductions in
U.S. forestry and agriculture over the next several
decades and beyond. It reports these reductions as
changes from baseline trends, starting in 2010 and
projected out 100 years to 2110. The report employs
the Forest and Agriculture Sector Optimization
Model with Greenhouse Gases (FASOMGHG).
FASOMGHG is a partial equilibrium economic
model of the U.S. forest and agriculture sectors,
with land use competition between them, and
linkages to international trade. FASOMGHG
includes most major GHG mitigation options in
U.S. forestry and agriculture; accounts for changes
in CO2, CH4, and N2O from most activities; and
tracks carbon sequestration and carbon losses over
time. It also projects a dynamic baseline and reports
all additional GHG mitigation as changes from
that baseline. FASOMGHG tracks five forest
product categories and over 2,000 production
possibilities for field crops, livestock, and biofuels
for private lands in the conterminous United States
broken into 11 regions. Public lands are not included.
FASOMGHG evaluates the joint economic and
biophysical effects of a range of GHG mitigation
scenarios, under which costs, mitigation levels,
eligible activities, and GHG coverage may vary.
The six scenarios evaluated in this report are
constant GHG prices, rising GHG prices, fixed
national mitigation levels, inclusion of selected
mitigation activities only, incentive payments for
1 A tonne is a metric ton, which equals one megagram (Mg). 1 tonne CO2 = 0.27 tonnes of carbon. 1 tonne of carbon = 3.67 tonnes
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
ES-1
-------
EXECUTIVE SUMMARY
CO2 only, and payments on a per-acre versus
per-tonne basis. GHG mitigation incentives are
estimated by dollars per tonne of CO2 equivalent
($/t CO2 Eq.) payments for four of the six scenarios
above. The model and analysis cover the 100 years
from 2010 to 2110, but three focus dates are high-
lighted: 2015, 2025, and 2055. FASOMGHG's
standard GHG accounting and payment approach
is a comprehensive, pay-as-you-go system, for all
applicable GHGs and activities over time.
The analysis reported here is unique from other
studies conducted on forestry and agricultural
mitigation options on a number of fronts. First,
the range of covered activities across the sectors
is wide. Most comparable studies look at just one
of the sectors or at one or a small subset of activi-
ties within each sector, while this report examines
a fairly comprehensive set of activities across the
two sectors covering a vast majority of all GHG
effects. Of particular note are the inclusions of
biofuels and non-CO2 mitigation options in agri-
culture. Second, the intertemporal dynamics of
the economic and biophysical systems within
FASOMGHG allow for an accounting of mitigation
over time and by region, and for quantification of
leakage effects that other studies generally have
not produced. And third, the inclusion of non-
GHG co-effects allows insights into the multiple
environmental and economic tradeoffs that pertain
to GHG mitigation in these sectors.
Highlights of the analysis include the following:
CHG reduction incentives can generate
substantial mitigation from the U.S. forest
and agriculture sectors especially in the first
few decades. Total national mitigation annually is
estimated to average almost 630 Tg CO2/yr (170 Tg
C) in the first decade and 655 Tg CO2/yr (180 Tg C)
by 2025, under one of the moderate GHG prices
considered ($15 t/CO2 Eq, or $55/t C, remaining
constant over time). Mitigation then declines to
about 85 Tg CO2/yr (23 Tg C) by 2055. The rate of
annual mitigation (i.e., occurring in a given year)
declines over time, as the result of saturating
carbon sequestration (to a new equilibrium) in
forestry and agriculture and carbon losses after
timber harvesting. Cumulative GHG mitigation
(i.e., achieved in the years up to a given year),
however, steadily increases for constant price
scenarios.
If GHG prices rise over time, however, GHG
mitigation is shown to start low and increase
over time. Farmers and foresters who want to
optimize their returns from any GHG payments
are assumed to know that GHG prices will rise in
future decades and may delay mitigation practices
until prices rise. The mitigation timing results,
however, are sensitive to the FASOMGHG model's
assumptions about landowner knowledge of future
price behavior, known as perfect foresight.
The optimal portfolio and timing of mitigation
strategies are affected by the GHG price levels.
At relatively low GHG prices (<$5/t CO2 Eq.) and
in early years, carbon sequestration in agricultural
soils and carbon sequestration in forest manage-
ment (i.e., harvest and regrowth practices) are the
dominant mitigation strategies. Afforestation
becomes the leading strategy at middle to higher
prices (>$15/t CO2 Eq.) in the early to middle years
to 2050, but both afforestation and sequestration in
agricultural soils get reversed by 2055, because of
carbon saturation, harvesting, and practice rever-
sion. Biofuels dominate the portfolio at the highest
prices ($30 and $50/t CO2 Eq.) and in later years
beyond 2050.
Agricultural CH4 and N2O mitigation is
a relatively small but steady part of the
mitigation portfolio. Biofuels and agricultural
CH4 and N2O mitigation are permanent emissions
reductions (i.e., they do not face the risk of GHG
benefit reversal).
Mitigation potential is likely to have a regional,
uneven distribution. The South-Central, Corn
Belt, and Southeast regions possess the largest
competitive potential to generate GHG mitigation,
while the Rockies, Southwest, and Pacific Coast
regions generate the least mitigation. Forest
management in the South-Central region generates
the most GHG mitigation, followed by agricultural
soil carbon sequestration in the Corn Belt, Lake
ES-2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
EXECUTIVE SUMMARY
States, and Plains, in low, constant price scenarios.
Afforestation in the South-Central and Corn Belt
regions is dominant at higher price scenarios.
Biofuels become a significant part of the mitigation
portfolio at high prices and occur primarily in the
Northeast, Southeast, and South-Central regions.
If a national GHG mitigation quantity in a
given year is an objective, but economic
incentives do not continue after that date,
then carbon sequestered in previous decades
is likely to be reversed. Landowners return to
other, more economically attractive land manage-
ment choices when GHG incentives disappear.
Leakage of GHG benefits from management
activities in one region to other regions may
be significant in scenarios where only selected
activities (e.g., afforestation) are eligible for
inclusion in a mitigation scheme. This leakage
may vary by activity, by region, and over time.
Agricultural activities, including soil carbon
sequestration, appear to have minimal leakage,
however (less than 6 percent).
Large changes in land use and production due
to mitigation activities can have substantial
non-GHG environmental co-effects. Even a low
GHG price (e.g., $5/tonne) can induce changes in
tillage practices and promote agricultural soil
carbon sequestration at a significant scale. Tillage
practice changes also reduce erosion and nutrient
run-off into waterways as a co-benefit, but can lead
to a modest increase in pesticide use as a co-cost.
Taking environmental co-effects into consideration
could affect the relative attractiveness of compet-
ing mitigation options. In general, the more
aggressive the mitigation action, the more likely
that co-effects may factor into the net benefits of
GHG mitigation.
Several key issues related to the design of an
incentive system can affect the magnitude,
timing, and duration of GHG benefits and cost.
These issues include if, and how, baseline setting,
leakage of GHG benefits, and the risk of reversal of
carbon management mitigation are addressed.
Another key issue is how mitigation is quantified
and reported. Use of cumulative mitigation (i.e.,
total mitigation to some future date) rather than
annual mitigation (i.e., in a given year) may more
accurately summarize the net GHG contribution
of forest or soil carbon management activities that
face some risk of reversal. Other considerations
include which activities are eligible for inclusion,
payment options (per acre versus per tonne), and
the potential adjustment of mitigation benefits to
account for reversal risk, leakage, and baseline
additionality.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
ES-3
-------
EXECUTIVE SUMMARY
ES-4 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 1
Introduction
Forestry and agricultural activities are widely
recognized as potential greenhouse gas
(GHG) mitigation options. Activities in
forestry and agriculture can reduce and avoid the
atmospheric buildup of the three most prevalent
GHGs directly emitted by human actions: carbon
dioxide (CO2), methane (CH4), and nitrous oxide
(N2O). CO2 is the gaseous form of carbon bound
with oxygen atoms.
The removal of atmospheric CO2 through seques-
tration in carbon "sinks" is a mitigation option in
forestry and agriculture that has received particu-
lar attention. Sequestration is the process of
increasing the carbon content of a carbon pool
other than the atmosphere (IPCC 2000). Terrestrial
carbon pools include tree biomass (roughly 50
percent carbon), soils, and wood products. A
carbon pool is a net sink if, over a certain time
interval, more carbon is flowing into the pool than
is flowing out of the pool. Likewise, a carbon pool
can be a net source of CO2 emissions if less carbon
is flowing into the pool than is flowing out of the
pool (IPCC 2000).
The forest and agriculture sectors can therefore
act as either sources or sinks of CO2 emissions.
Agriculture (including croplands and livestock)
is a particularly large source of CH4 and N2O
emissions. Globally, land-use change, primarily
tropical deforestation, accounts for approximately
20 percent of the world's annual, anthropogenic
CO2 emissions (IPCC 2000). An even greater
amount of atmospheric CO2 is removed by forests
than is emitted by land-use change, such that the
net global terrestrial sink (sink minus source)
offsets approximately 11 percent of the world's CO2
emissions due to fossil fuel combustion (IPCC
2000). Meanwhile, agriculture accounts for ap-
proximately 50 percent of global anthropogenic
CH4 emissions and 85 percent of global N2O
emissions (Scheehle and Kruger in press). CH4
and N2O are relatively potent greenhouse gases
and can be placed on a comparable climatic basis
with CO2 through a Global Warming Potential
(GWP) factor (see Box 1-1).
Box 1-1: Relative Global Warming Potential
of Non-CO2 Gases
The Global Warming Potential (GWP) compares the
relative ability of each GHG to trap heat in the
atmosphere over a certain time frame. Per IPCC
(1996) guidelines, CO2 is the reference gas and thus
has a GWP of 1. Based on a time frame of 100 years,
the GWP of CH4 is 21, implying that a ton of methane
is 21 times more potent than a ton of CO2. The GWP
for N2O is 310. These values can be further trans-
formed from CO2 to carbon equivalent by dividing
by 3.67, the mass ratio of CO2 to C.
Note that GWPs from the IPCC Third Assessment
Report (2001) are not used in this report because
international GHG reporting guidelines are still based
on the 1996 IPCC Second Assessment Report.
In the United States, forest and agricultural
lands also comprise a net carbon sink. Removal of
atmospheric CO2 through sequestration is greater
than CO2 emissions through events such as forest
harvests, land conversions or other uses, or fire.
The U.S. carbon sink—over 90 percent of which
occurs on forest lands—currently offsets 12
percent of U.S. GHG emissions from all sectors
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
1-1
-------
CHAPTER 1 • INTRODUCTION
of the economy (EPA 2005; Figure 1-1). Agriculture
accounts for about 30 percent of all CH4 emissions
and 72 percent of all N2O emissions in the United
States (op cit). Taken together, agricultural CH4
and N2O emissions are responsible for about 6
percent of all U.S. GHG emissions, expressed on
a GWP-weighted CO2 equivalent basis (op cit).
Key individual GHG mitigation options in U.S.
forestry and agriculture include
• afforestation (tree planting);
• forest management, including silviculture,
harvests, and forest preservation;
• agricultural soil carbon sequestration (primarily
through changes in cropland tillage practices);
• fossil fuel use reduction associated with altered
practices in agriculture;
• agricultural CH4 and N2O emission reduction
(through a variety of modifications to livestock
management and fertilizer applications); and
• biofuel offsets of fossil fuels (derived from
bioenergy crops such as switchgrass).
These options generally fall into three categories
(see IPCC [2001, 2000]): 1) options that avoid CO2
emissions by preserving existing pools or sinks
of carbon in tree biomass and soils (e.g., forest
preservation), 2) options that enhance the removal
of atmospheric CO2 (sinks) through sequestration
(e.g., afforestation), and 3) options that directly
reduce fossil fuel-related CO2 or CH4 and N2O
emissions (e.g., biofuels and reduced fertilizer
use). Chapter 2 discusses the individual mitigation
options in greater detail.
Forestry and agricultural activities that either
preserve or enhance carbon sinks exhibit unique
and important features compared to mitigation
options that directly reduce fossil fuel-related CO2
or CH4 and N2O emissions. Two distinguishing
characteristics are the saturation over time of
carbon sequestration in vegetative biomass and
soils, as a new equilibrium is reached for a given
level of inputs, and the potential reversibility, or
re-release, back to the atmosphere of sequestered
carbon through natural or anthropogenic distur-
bances (e.g., tillage, or fire). The reversibility of
Figure 1 1: Forestry and Agriculture Net Contribution to GHG Emissions in the United States, 2003a
O
O
01
8,000
7,000 -
6,000 -
5,000 -
4,000 -
3,000 -
2,000 -
1,000 -
0
-1,000 -
-2,000
6,900
5,842
-7
-753
-828
Ag Soil Forest Total U.S.
Sequestration Sequestration Sequestration
Ag N2O and Total U.S. CO2
CH.
(all sectors)
Total U.S.
GHGs
(all sectors)
a Total agriculture and forestry sequestration also includes urban trees and landfilled yard trimmings and food scraps. Negative
values represent a sink, positive values a source.
Source: EPA (2005).
1-2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 1 • INTRODUCTION
carbon sequestration benefits is often referred to
as the duration or permanence issue. Analyses
presented in the report highlight the implications
of saturation and reversibility of carbon sequestra-
tion in forestry and agriculture.
Purpose and Approach of this Report
This report aims to assess the GHG mitigation
potential from forestry and agriculture in the
United States over the next several decades, out to
the 2050s, and in some cases beyond.
More specifically, the report aims to examine the
following questions:
• What is the total GHG mitigation potential of
the full suite of forestry and agricultural activi-
ties over time and at different costs?
• How does the portfolio of forestry and agricul-
tural activities change over time and at different
levels of GHG reduction incentives (or "GHG
prices")?
• What is the regional distribution of GHG
mitigation opportunities within the United
States?
• How does the portfolio of activities, time profile,
and regional distribution change across scenari-
os that reflect constant prices for GHG mitiga-
tion, rising prices, and fixed mitigation levels?
• What are the implications of carbon saturation
and reversibility (or duration)?
• How do leakage and other implementation
issues affect GHG mitigation benefits?
• What are some of the non-GHG environmental
co-effects of GHG mitigation activities?
• What appear to be the top mitigation options,
nationally and regionally, taking GHG, econom-
ic, implementation, and other environmental
factors into account?
The analysis uses the Forest and Agriculture
Sector Optimization Model with Greenhouse
Gases (FASOMGHG) to examine these questions.
FASOMGHG is a partial equilibrium economic
model with comprehensive GHG accounting of the
forest and agriculture sectors of the U.S. economy,
linked to the rest of the world by international
trade linkages. FASOMGHG can gauge the nation-
al aggregate response to GHG incentives (prices
or GHG mitigation targets) and identify the
most cost-effective mitigation opportunities at
the national and regional levels. FASOMGHG
can examine various scenarios with different
approaches to achieving GHG mitigation (e.g.,
where all forestry and agricultural activities are
included, where individual activities are included,
or where all or individual GHGs are included).
All reported GHG mitigation activities in
FASOMGHG occur as changes from a business-
as-usual or baseline trajectory of carbon seques-
tration rates, GHG emissions, and economic
activity in U.S. forestry and agriculture over
time. Thus, the mitigation results reported here are
additional to projected baseline activity and GHG
emission or sequestration levels. FASOMGHG also
reports some non-GHG environmental co-effects
(such as changes in nonpoint loadings of nitrogen
and phosphorous from agriculture) for a more
complete analysis of mitigation outcomes.
Organization of Report
This report is organized as follows:
• Chapter 2 describes the GHG mitigation
options in U.S. forestry and agriculture repre-
sented in the FASOMGHG model, as well as
some others not explicitly modeled for this
report.
• Chapter 3 presents the modeling framework
of FASOMGHG and the model's projected
baseline (with a brief comparison to other
baseline studies), against which all mitigation
estimates in subsequent chapters are reported.
• Chapter 4 presents GHG mitigation results
for the full suite of forestry and agricultural
activities. Scenarios include a range of constant
and rising GHG price incentives over time.
Regional GHG mitigation results for these
scenarios are presented as well.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
1-3
-------
CHAPTER 1 • INTRODUCTION
Chapter 5 presents GHG mitigation results for
the following selective scenarios: 1) three fixed
GHG mitigation levels, 2) selection of individual
or subsets of forestry and agricultural activities,
and 3) addressing of CO2 reductions only (versus
all GHGs).
Chapter 6 evaluates some implications of taking
activity-specific mitigation approaches and
different payment methods. The chapter also
presents estimates of the potential for "leakage,"
or the shifting of emissions to activities not
subject to incentives.
Chapter 7 provides more detail on the
non-GHG environmental co-effects of GHG
mitigation activities.
Chapter 8 concludes the report by highlighting
the report's key findings and the insights they
hold for the realization of GHG mitigation
potential in forestry and agriculture.
1-4 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 2
Greenhouse Gas Mitigation Options
in U.S. Forestry and Agriculture
Chapter 2 Summary
GHG mitigation opportunities in forestry and agriculture include afforestation (tree planting), forest
management (e.g., altering harvest schedules or management inputs), forest preservation, agricul-
tural soil tillage practices, grassland conversion, grazing management, riparian buffers, biofuel
substitutes, fertilization management, and livestock and manure management. Each of these oppor-
tunities is described, with emphasis on their ability to avoid, sequester, and/or reduce CO2, CH4,
and N2O emissions. Sequestration activities can enhance and preserve carbon sinks and include
afforestation, forest management, and agricultural soil tillage practices. Agricultural sources of CH4,
N2O, and fossil fuel CO2 can be reduced through changes in fertilizer applications and livestock and
manure management. CO2 emissions can be offset through biofuels, such as switchgrass and short-
rotation tree species, which can be grown and used instead of fossil fuels to generate electricity.
This chapter also considers the unique time dynamics and accounting issues of carbon seques-
tration options: saturation (or equilibrium level) of carbon sequestration overtime, potential revers-
ibility of carbon benefits, and fate of carbon stored in products after forest harvests. In contrast,
agricultural non-CO2, fossil fuel CO2, and biofuel options do not exhibit saturation or reversibility
and are therefore generally considered permanent. Most mitigation opportunities described in this
chapter are included in the analyses described in later chapters.
Forestry and agricultural activities can help
reduce and avoid the atmospheric buildup
of CO2, CH4, and N2O in a number of ways.
Atmospheric CO2 can be removed and sequestered
in tree biomass and soils, which can act as carbon
sinks. Carbon stored in tree biomass and soils can
be protected and preserved to avoid CO2 releases
to the atmosphere. Emissions of CO2 can be
avoided by reducing the use of energy-intensive
inputs or by using biofuels, produced in the forest
and agriculture sectors, instead of fossil fuels to
produce energy. And agricultural CH4 and N2O
emissions can be directly reduced by modifying
livestock management and fertilizer applications.
This chapter discusses the key forestry and
agricultural mitigation options that either avoid,
sequester, and/or reduce CO2, CH4, and N2O. This
chapter also discusses important issues related to
the reversibility or permanence of forestry and
agricultural options involving carbon sinks. The
chapter presents the individual mitigation options
as activities undertaken by landowners at the farm
or forest-stand level. Subsequent chapters charac-
terize the extent to which these mitigation options
can be brought about by economic incentives
operating at a nationally or regionally aggregated
level. Examples of such incentives currently in
place include government programs such as the
Farm Bill, or voluntary GHG registries.
Carbon Sequestration
A number of practices within the forest and
agriculture sectors can mitigate the atmospheric
build-up of GHGs by removing CO2 from the
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
2-1
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
atmosphere and then storing it in forest and agro-
ecosystems at a rate greater than its release back
to the atmosphere through human and natural
disturbances. These carbon sequestration activities
can take on a variety of forms as discussed below.
Afforestation
Afforestation can be defined broadly as the
establishment of trees on lands that were without
trees for some period of time. Differing interpreta-
tions of this time period will dictate whether the
establishment of forest cover is considered to
represent afforestation or reforestation. The
Intergovernmental Panel on Climate Change
(IPCC) defines afforestation as the planting of
new forests on lands that, historically, have not
contained forests (IPCC 2000).
Reforestation often refers to the reestablishment
of forest after a harvest in the United States. This
report treats reforestation, or changes in the
harvest-regeneration cycle, as part of "forest
management," discussed below. FASOMGHG
models afforestation separately, but reforestation
is embedded within the broader activity of forest
management in FASOMGHG and not treated
separately.
Afforestation enhances carbon sequestration
because land is allocated away from uses with
relatively low carbon storage potential (e.g.,
conventional crop agriculture) to forest cover with
higher carbon storage potential. Carbon accumu-
lates in forest soils and biomass, the latter both
below ground in the form of roots and above
ground in stem, branches, and leaves. The rate of
carbon accumulation for afforestation varies and
depends on the newly planted tree species, climate,
soil type, management, and other site-specific
characteristics (e.g., 2.2 to 9.5 tonnes of CO2 per
acre per year, as reported by Birdsey [1996]; see
Table 2-1). As a carbon sequestration activity,
afforestation primarily affects atmospheric CO2.
The movement of land from agricultural use to
forest also generally leads to a reduction in the
various GHG emissions from agriculture, as
described below. Most recent afforestation in the
United States has occurred on pasturelands, where
from 1982 to 1997 over 14 million acres were
converted to forest cover (USDA NRCS 2000).
Forest Management
Forest management has traditionally focused on
maximizing the value of harvested commercial
timber over time. However, forests also can be
managed to enhance carbon sequestration, via
silvicultural practices or conservation of standing
stocks. A managed forest will consist of one or
several tree species in stands, and the mix can be
designed so that the trees aid one another to ensure
the fastest and most efficient biomass growth
and thus higher sequestration potential. The
landowner may choose to plant a moderately fast-
growing species to accumulate timber (and carbon)
faster; he or she may also use practices such as
fertilization, controlled burning, and thinning to
increase forest and carbon productivity.
Managed forests pass through multiple stand ages
ranging from stand establishment to harvest. In a
forest managed for timber production, the optimal
harvest age is the time when the value of the
additional timber growth obtained by delaying
the harvest further is overtaken by the opportunity
cost of the delay. Traditional forest rotation lengths
vary by region and species type. The nonindustrial
private forests (NIPF) of the southern United States
are commonly managed with softwood or mixed
species on a rotation of approximately 25 to 35
years or more. Rotations in commercial forestry,
as practiced on forest industry-owned lands or
very intensively managed NIPF lands, may be as
short as half the length of the more typical NIPF
rotation. The forest rotations of the western United
States tend to be longer (between 45 and 60 years),
because they consist of species that culminate
growth at a later age. The varying rotation lengths
allow for the production of multiple forest products
including smaller-diameter pulpwood and larger-
diameter sawtimber.
When carbon is considered a forest output, the
value of delaying the rotation is higher because
carbon accumulates as the trees grow (van Kooten,
Binkley, and Delcourt 1995, Murray 2000). Thus,
forest managers can enhance carbon sequestration
2-2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
Table 2-1: Representative Carbon Sequestration Rates and Saturation Periods for Key Agriculture,
Land-Use Change, and Forestry Practices
Activity
Representative Carbon
Sequestration Rate in U.S.
(Tonnes of CO2 per acre per year,
unless otherwise indicated)
Time Over which Sequestration May
Occur before Saturating
(Assuming no disturbance, harvest,
or interruption of practice)
References
Afforestation3
Reforestation0
Avoided deforestation
Changes in
forest management
Reduced tillage
on croplands9
Changes in grazing
management
Cropland conversion
to grassland
Riparian buffers (nonforest)
Biofuel substitutes
for fossil fuels
2.2-
1.1 -
83.7-
2.1
0.6
0
0.07
0.9
0.4
4.8-
-9.5b
-7.7d
•172.19
-3.1f
-1.1
.7h
-1.91
-1.9'
-1.0
-5.5k
90 -120+ years
90 -120+ years
N.A.
If wood products included in accounting,
saturation does not necessarily occur if
carbon continuously flows into products
15 -20 years
25 - 50 years
25 - 50 years
Not calculated
Not calculated
Saturation does not occur if fossil fuel
emissions are continuously offset
Birdsey (1996)
Birdsey (1996)
U.S. Government (2000)
Row (1996)
West and Post (2002)
Lai etal. (1998)
Folletetal. (2001)
Eve et al. (2000)
Lai etal. (1998)
Lai etal. (1998)
Note: Any associated changes in emissions of CH4 and N2O or—except for biofuels—fossil fuel CO2 are not included.
a Values are for average management of forest after being established on previous croplands or pasture.
b Values calculated over 120-year period. Low value is for spruce-fir forest type in Lake States; high value for Douglas fir on Pacific
Coast. Soil carbon accumulation included in estimate.
c Values are for average management of forest established after clearcut harvest.
d Values calculated over 120-year period. Low value is for Douglas fir in Rocky Mountains; high value for Douglas fir in Pacific
Northwest. No accumulation in soil carbon is assumed.
9 Values represent the assumed CO2 loss avoided in a single year (not strictly comparable to annual estimates from other options).
Low and high national annual average per acre estimates based on acres deforested from National Resource Inventory (NRI) data
and carbon stock decline from the FORCARB model, from 1990 to 1997.
f Selected example calculated over 100 years. Low value represents change from unmanaged forest to plantations for pine-
hardwood in the mid-South; high value is change from unmanaged forest to red pine plantations for aspen in the Lake States.
9 Both West and Post and Lai et al. estimates here include only conversion from conventional to no till. Estimates do not include
fluxes of other associated GHGs.
h Tillage rates vary, but this value represents a central estimate by Lai et al. for no-till, mulch till, and ridge till.
1 Low-end estimate is for improved rangeland management; high-end estimate is for intensified grazing management on pastures,
which includes the return of plant-derived carbon and nutrients to the soil as feces.
' Assumed that carbon sequestration rates are same as average rates estimated for lands under the USDA Conservation Reserve
Program (CRP).
k Assumes growth of short-rotation woody crops and herbaceous energy crops, and an energy substitution factor of 0.65 to 0.75.
Potential for changes in other GHG emissions not included.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
2-3
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
by extending the harvest age of the managed
forests. Over time, a new and higher carbon
equilibrium will be reached. Carbon sequestration
rates due to forest management practices vary
depending on the practice itself, tree species,
climate, topography, and soil type (e.g., 2.1 to 3.11
CO2/acre/year as reported by Row (1996); see
Table 2-1).
When a forest is harvested, some carbon is imme-
diately released to the atmosphere via the logging
operation or milling process (about one-half or
two-thirds is emitted at or near the time of harvest,
depending on the product and region), but some
is tied up in wood products for a number of years.
Carbon from wood products may be released to the
atmosphere many years in the future as the wood
products decompose, the timing of which will
depend on whether the products are short-lived
(e.g., paper) or long-lived (e.g., housing lumber),
and whether those products are discarded in
landfills. The carbon sequestration and emissions
that result from the harvest-regeneration cycle,
including the wood products pool, are captured
in the analyses presented later in the report.
Forest management primarily affects carbon pools
and associated atmospheric CO2, rather than fossil
fuel CO2 and non-CO2 emissions. Although it uses
equipment to establish, cultivate, and harvest
stands of trees, forestry is less energy-intensive
than agriculture because the management inter-
ventions are spread out episodically over time—a
handful of interventions at most over 20 to 50 years
for managed stands, less for stands that remain
unmanaged. Therefore, there is limited ability to
reduce energy-related CO2 emissions in forestry.
N2O can be generated from forest fertilizer
applications. However, relatively few forested acres
receive fertilizer applications in a given year, so
1 N2O emissions associated with fertilization of forest soils are estimated to be 0.4 Tg CO2 Eq. in the Inventory of U.S. Greenhouse
Gas Emissions and Sinks: 1990-2003 (EPA 2005). These emissions are not included in the analyses presented in later chapters.
According to EPA (2005), the rate of fertilizer application for the area of forests that receives fertilizer in any given year is
relatively high. However, average annual applications are quite low (inferred by dividing all forestland by the amount of
fertilizer added to forests in a given year).
2 A mature forest, however, is not a static or unchanging carbon source; it is just that the net rate of sequestration is on average
unchanging. But some studies suggest that even very old forested stands continue to sequester carbon (Lugo and Brown 1986,
Phillips et al. 1998, Phillips et al. 2002a).
the aggregate effect of forestry on N2O emissions
is quite small.1
A form of forest management that can avoid CO2
emissions is forest preservation, sometimes referred
to as forest protection or a harvest set-aside. This
entails adopting a management regime that does
not involve harvesting. Although CO2 emissions
from harvesting may be avoided, the enhancement
of carbon storage will cease when the forest meets
its biophysical equilibrium—when carbon inputs
equal carbon outputs. The carbon stock then
essentially becomes a static pool.2 Preservation of
this form foregoes the option to replace a steady-
state forest with a net-sequestering young forest.
However, as shown in Harmon et al. (1990) after
timber harvests in the Pacific Northwest, the on-
site carbon declines significantly and it takes over
200 years for a newly reforested area to attain the
storage capacity of an old growth forest.
The GHG benefits of reducing or avoiding deforesta-
tion in many ways simply mirror those from
afforestation. However, there may be significant
differences in the timing of GHG effects. Under
afforestation, it takes decades for carbon to accu-
mulate in forest soils and biomass. The process of
deforestation—clearing forestland for another use
—may release a substantial amount of carbon into
the atmosphere rapidly upon the time of harvest.
Although some carbon may be transferred off-site
in the form of harvested wood products, a substan-
tial portion is released immediately in harvesting
and manufacturing (Skog and Nicholson 2000), on
the order of, say, 150 to 800 t CO2/acre.
The USDA's Natural Resources Inventory (NRI)
shows that 5.7 percent of the private forested land
base in the United States was deforested between
the years 1982 and 1997 (USDA NRCS 2000), at an
2-4
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
average annual rate of 241,147 acres per year. The
primary conversion of forestland was to pasture
and developed lands.
Avoiding or reducing deforestation does not
necessarily imply that harvests will never occur.
Rather, land can be retained in forested use and
still be managed to produce timber through
periodic harvesting. The process of eliminating
harvests altogether is referred to as forest preser-
vation or forest protection, as discussed above.
Agricultural Soil Carbon Sequestration
Croplands often emit CO2 as a result of conven-
tional tillage practices and other soil disturbances.
Soils containing organic material that would
otherwise be protected by vegetative cover are
exposed through conventional tillage practices and
become susceptible to decomposition. Frequent or
intense tillage breaks down soil macroaggregates,
thereby enhancing the exposure of carbon to
microbial activity. This added soil exposure also
enhances decomposition by raising the soil tem-
perature (Lai et al. 1998). Adopting conservation
tillage practices, changing the overall land and
crop management, modifying cropping intensity,
or retiring marginal lands from production can
reduce or eliminate this exposure, thus reducing
or eliminating the associated CO2 emissions.
Given widespread adoption of the management
options discussed here, agricultural soils may be
able to contribute more than a reduction in emis-
sions; they have the potential to become a net sink
of CO2. These options are discussed briefly below.
In the United States, conservation tillage is typi-
cally defined as any tillage system that maintains
at least 30 percent of ground covered by crop
residue after planting (CTIC 1994). Conservation
tillage eliminates one or several of the practices
associated with conventional tillage, such as
turning soils over with a moldboard plow and
mixing soils with a disc plow (Lai et al. 1998).
Conservation tillage practices, including no till,
ridge till, and minimum till, allow crop residues
to remain on the soil surface as protection against
erosion.
Current estimates for CO2 gains from conservation
tillage range from about 0.6 to 1.1 t/CO2/acre/yr,
with differences in the estimated saturation period
(West and Post 2002, Lai et al. 1998). A compilation
of study results by West and Post (2002) suggests
that soil carbon accumulation after adoption of
conservation tillage typically occurs for periods
of 15 to 20 years and then returns to a soil carbon
steady state with no additional gains in carbon.
Studies suggest that agricultural soils in the
United States, on aggregate, have not reached a
biophysical saturation point (IPCC 2000, Donigian
et al. 1995, Kern and Johnson 1993). Further
information on carbon saturation and reversal
issues is provided below.
A final option aimed at reducing the potential
decomposition of organic material is the retire-
ment of economically marginal lands from produc-
tion. Removing these lands from production can
reduce CO2 emissions, as well as N2O emissions
associated with fertilizer applications. Depending
on the new land cover of these retired lands, they
can become a carbon sink. Lands are often retired
through federal programs such as the USDA
Conservation Reserve Program (CRP).
Grassland Conversion
Grassland conversion refers to converting existing
cropland to grasslands or pasture. Because there is
continuous vegetative cover, the retention of soil
carbon is higher than that for conventionally tilled
cropland. Grassland conversion often involves
cropland needing conservation treatments and
may be part of a conservation program, such as
CRP. Sequestration from this activity can vary
from about 0.9 to 1.9 t CO2/acre/yr (Eve et al. 2000,
Table 2-1).
Grazing Management
While expanding grassland area can enhance
carbon storage, further sequestration may be
possible from improving the way grasslands are
used for livestock grazing. Sequestration can be
enhanced by increasing the quantity and quality of
forages on pastures and native rangelands and by
reducing carbon losses through the degradation
process, thereby retaining higher soil carbon
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
2-5
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
stocks (IPCC 2000). The range of mitigation
estimates for grazing practices is wide, and the
applicability of these numbers to the United States
is a topic of ongoing research.
Grazing management practices can have multiple
GHG effects. For instance, the quality of forage
can affect livestock digestion processes and the
amount of CH4 that is emitted through enteric
fermentation. Additionally, if nutrient inputs, in
particular nitrogen-based fertilizers, are needed
to enhance forage stocks, this can generate N2O
emissions post-application. The CH4 and N2O
implications of livestock practices are addressed
in more detail below.
Riparian Buffers
The establishment of riparian buffers can be
viewed as a special case of either afforestation,
forest management, or grassland conversion and
thus fall under either forestry or agriculture. These
practices are of particular interest because of their
potential water quality co-benefits. Riparian
buffers involve the establishment or maintenance
of coarse vegetative land cover (trees, brush,
grasses, or some mixture) on land near rivers,
streams, and other water bodies. These actions
are often focused around areas being cultivated
or developed and used to filter the runoff of
sediment, nutrients, chemicals, and other com-
pounds that may impair water quality. Local, state,
or federal government or private company guide-
lines often mandate that existing riparian buffers
be left intact during timber harvests. Establishing
or protecting these buffers can sequester CO2 in
the soil from the accumulation of organic material
and in vegetative biomass if the buffer is planted
or vegetation migrates into the area. This option
also reduces baseline emissions from agriculture
if the total cultivated area declines.
In 1997, a total of 199,600 acres of field borders and
filter strips were in place on cropland, and a total
of 1.6 million acres of grassed waterways existed
(Uri 1997).
GHG Emissions Reduction Options
This section presents the agricultural mitigation
options that can directly reduce CO2, CH4, and
N2O emissions, separate from the carbon seques-
tration options discussed above. CO2 emission
reduction options are discussed first; then the
section addresses options to reduce non-CO2 gases.
Reduction of CO2 Emissions from Fossil
Fuel Use
The main direct source of CO2 emissions from
U.S. agriculture is on-farm fuel use, although there
are upstream releases related to the manufacture
of equipment, fertilizer, and other agricultural
inputs. Changes in practices that reduce the use of
energy-intensive inputs can reduce CO2 emissions
from this sector. In the analysis presented in
subsequent chapters, the CO2 emissions captured
because of agricultural management changes
include emissions from direct use of fossil fuels
in farm equipment, water pumping, and grain
drying and fossil fuel use in fertilizer and pesticide
production. For the purposes of this report, these
emission reductions are associated with agricul-
tural-sector activity, but other reports (e.g., annual
EPA Inventory of U.S. Greenhouse Gas Emissions and
Sinks) may consider these emissions associated
with the energy or manufacturing sector.3
Reduction of Non-CO2 GHG Emissions
Agriculture is a major source of non-CO2 GHGs
emissions, and the emissions can be reduced in
numerous ways through changes in management
practices. The GHGs of primary concern in the
agriculture sector are N2O and CH4. These agricul-
tural gases account for 433 Tg CO2 Eq./year or over
6 percent of total U.S. GHG emissions (EPA 2005).
Figure 2-1 displays the relative contribution of
these activities and compares them to total U.S.
GHG emissions. The relative potency of N2O and
CH4 as climate change gases is greater than CO2
on a per-unit basis (see Box 1-1 in Chapter 1).
3 Please note that this report does not consider emissions from fossil fuel use in the forestry sector because of insufficient data on
these emissions.
2-6
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
Figure 2 1: Agricultural Non CO2 Emissions by Source Relative to All Other GHG Emissions
All Other
GHG
Emissions
(94%)
Rice Cultivation (1.6%)
Manure Management (13%)
Enteric Fermentation (26.5%)
Ag. Soil Management (58.5%)
Other (0.3%)
Source: EPA (2005).
N2O emissions from agriculture account for just
over 270 Tg CO2 Eq./year or 63 percent of agricul-
tural non-CO2 emissions. Agricultural N2O is
largely tied to fertilizer application, nitrogen-fixing
plants such as legumes, and manure emissions.
Therefore, reductions can be accomplished by
reducing nitrogen-based fertilizer applications,
using nitrogen inhibitors, improving nitrogen
nutrient management, altering crop mix, and
reducing nitrogen content of animal feeds (McCarl
and Schneider 2000). Economic incentives to
reduce GHGs can alter the relative price of inputs
and management practices that generate non-CO2
emissions. The economic model used in this report
accounts for these changes in prices (costs) and
modifies practices and reduces emissions accord-
ingly in the analyses that follow.
CH4 emissions account for 161.4 Tg CO2 Eq.
per year or 37 percent of agricultural non-CO2
emissions and are due in large part to emissions
from livestock manure and enteric fermentation
in the digestive tracts of ruminant livestock (see
Table 2-2). Changes in feeding ratios and manure
management strategies can be undertaken to
reduce these emissions. Rice cultivation is also
a source of CH4 emissions, although less so in the
United States than in other parts of the world. CH4
uptake and emissions from cropland soils are not
well understood and are not included in the EPA
GHG inventory reports or in this analysis. The
following sections outline four major sources of
agricultural non-CO2 emissions and potential
mitigation options.
Table 2-2: Agricultural Non-CO2 Emissions by Source, 2003 (Tg CO2 Eq.)
Emission Source
CH4
N2O
Total Non-CO,
Agricultural soil management
Enteric fermentation
Manure management
Rice cultivation
Field burning of agricultural residues
—
115.0
39.1
6.9
0.8
253.5
—
17.5
—
0.4
253.5
115.0
56.6
6.9
1.2
Total emissions from agriculture
Source: EPA (2005).
161.8
271.5
433.2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
2-7
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
Agricultural Soil and Fertilization Management
N2O emissions are produced in soils through the
processes of nitrification (aerobic microbial oxida-
tion of ammonium [NH4] to nitrate [NO3]) and
denitrification (anaerobic microbial reduction of
nitrate to di-nitrogen [N2]). Agricultural soil N2O
emissions represent 58 percent (253.5 Tg CO2 Eq.)
of agricultural non-CO2 emissions (Table 2-2). The
application of nitrogen-based fertilizers to crop-
lands is a key determinant of N2O emissions,
because excess nitrogen not used by the plants is
subject to gaseous emissions, as well as leaching
and runoff. A viable mitigation option to reduce
soil N2O emissions is to adopt management
practices that ensure the most efficient use and
application of nitrogen-based fertilizer while
maintaining crop yields.
Enteric Fermentation
The primary source of CH4 emissions, which
represents 27 percent (115.0 Tg CO2 Eq.) of agricul-
tural non-CO2 emissions (Table 2-2), is ruminant
livestock and the microbial fermentation process of
feed in their digestive system (rumen). The amount
of CH4 emitted from an animal depends primarily
on the efficiency of the animal's digestive system,
which is determined by the animal's feed or diet.
Viable options are available for reducing CH4
emissions from enteric fermentation, because CH4
releases essentially represent wasted energy that
could otherwise be used to produce milk or beef.
Direct approaches attempt to increase the rumen
efficiency, thus reducing the amount of CH4
produced per unit of feed. Indirect options focus
on increasing animal productivity, reducing the
amount of CH4 emitted per unit of product (e.g.,
milk, beef). These direct and indirect approaches
include options for improving the feed-intake
efficiency (e.g., use of bovine somatotropin [bST]),
altering livestock management practices (e.g.,
elimination of stacker phase in beef production),
and using intensive grazing.
Manure Management
Livestock manure can produce both CH4 and N2O
emissions. The level of CH4 emissions depends on
the way the manure is handled and stored. In
many livestock operations in the United States,
animals are raised in confined areas, and their
manure is diverted to holding areas for further
management. CH4 is produced by the anaerobic
decomposition of manure that is stored in lagoons,
ponds, pits, or tanks. N2O is produced through
the nitrification and denitrification of the organic
nitrogen in livestock manure and urine. The
combined CH4 and N2O emissions from livestock
manure represent 13 percent (56.6 Tg CO2 Eq.)
of agricultural non-CO2 emissions (Table 2-2).
Anaerobic digesters that cover and capture the
CH4 emitted from collected manure, and poten-
tially used as an on-farm energy source, represent
a key mitigation option. The specific storage
system will determine the type of digester or
digestion process that will be applied to the
manure (e.g., plug and flow, unheated or heated
lagoon, complete mix). The emitted gas can either
be converted into electricity for use as an on-farm
energy source or consumed through flaring the
collected gas. In either case, CH4 is mitigated and
CO2 is released, but this option still remains a
viable option for net GHG reductions because the
GWP for CH4 is 21 times higher than CO2. Another
CH4 mitigation option allows for aerobic decompo-
sition of manure as a solid on pasture-, range-, or
paddock lands.
Rice Cultivation
Rice production under flooded conditions results
in CH4 emissions through the anaerobic decompo-
sition of organic matter in the fields. Approximately
90 percent of the world's harvested rice area is
grown under this management practice for some
period of time (Wassman et al. 2000). In the United
States, all rice is cultivated under flooded condi-
tions (EPA 2005), but rice CH4 accounts for less
than 2 percent (6.9 Tg CO2 Eq.) of U.S. agricultural
non-CO2 emissions (Table 2-2). Mitigation options
for rice CH4 include changes in water management
regime, the use of inorganic fertilizers, and differ-
ent cultivar selection. In the analyses presented
later in the report, rice CH4 is reduced through
decreases in rice acreage.
2-8
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
Biofuel Offsets of Fossil Fuels
Products from the forest and agriculture sectors
can mitigate GHGs by serving as substitutes for
fossil fuels or for products that depend on fossil
fuel combustion in their production. Though these
options do involve forest and agricultural carbon
sinks, the primary GHG benefits of these options
can generally be treated as equivalent to perma-
nent emission reductions.
A potential process for reducing atmospheric
CO2 is the cultivation of perennial grasses, short-
rotation woody crops, or traditional crops for
biofuel production. The production of these
alternative energy sources created from biomass
has the potential to reduce the use of fossil fuels
used in the power generation and transportation
sectors, the largest sources of CO2 emissions in
the United States.
The essential premise of biofuel as a means to
reduce GHGs is based on their renewability.
Biofuels, like fossil fuels, release GHGs when
burned for energy production. However, biofuels
are releasing GHGs (CO2) that have been removed
from the atmosphere through photosynthesis and
stored in biomass. In essence, the plants are
harvesting GHGs for use in energy production. In
a steady state of biofuel production and use, there
is little to no net addition to atmospheric GHG
concentrations. However, fossil fuel combustion
transfers carbon to the atmosphere that was stored
underground in coal, petroleum, or natural gas
reserves without replacing the fossil carbon stock
and thereby, on net, raises GHG concentrations.
Specific examples of biofuel options include using
forestry and agricultural residues and planting
dedicated energy crops such as switchgrass or
poplar to use as feedstock for electric power
generation. In 2002, biomass accounted for only 1
percent (37 billion kilowatt hours) of U.S. electric-
ity generation and is projected under baseline
conditions to remain at 1.3 percent of generation
(81 billion kilowatt hours) by 2025 (Energy Infor-
mation Administration [EIA] 2004). In analyses
presented later in this report, emission reductions
due to biofuels used in power generation result
from comparing net GHG emissions of coal-fired
plants to net GHG emissions of biomass-fired
plants. Using biofuels as a supplement to coal
in co-fired plants is also possible. Finally, corn can
be grown to produce ethanol as replacement for
liquid fossil fuels (though this latter option gener-
ates little GHG mitigation in this report's analysis).
Unique Time Dynamics of Carbon
Sequestration Options
Forestry and agriculture practices that preserve
and enhance carbon storage in soils and biomass
exhibit unique and important features compared
to mitigation activities in all sectors of the economy
that reduce fossil fuel CO2, CH4, N2O, and emis-
sions of other GHGs. The primary distinguishing
characteristics are mainly related to the unique
temporal dynamics of sequestration options.
Comprehensive GHG accounting of sequestration
options requires the inclusion of both sequestra-
tion and release of CO2 and sometimes CH4 and
N2O. This tracking needs to occur over long
timeframes both during normal land-use and
management practices and in mitigation activities.
Three fundamental factors need to be considered:
the slowdown or so-called saturation (or approach
to equilibrium) of sequestration rates, the potential
for reversal of carbon benefits if sequestered carbon
is re-released into the atmosphere at some future
point in time, and the fate of carbon in long-lived
products after the time of harvest. These issues of
carbon permanence are addressed briefly below
and more thoroughly again in Chapter 6.
"Saturation" of Carbon Sequestration
to Equilibrium
The amount of carbon that can be sequestered
in agricultural soils and forest ecosystems is
ultimately constrained by biophysical factors.
Once a sequestration activity such as afforestation
or crop tillage change takes place, the rate of the
ecosystem's carbon inputs exceeds the rate of its
carbon outputs, thereby leading to a net accumula-
tion of carbon stocks on-site. However, the bio-
physical processes evolve over time until the rate
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
2-9
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
of carbon output just equals the rate of carbon
inputs. At that point, the system has reached a
new carbon equilibrium, and no net carbon stock
accumulations can be expected beyond that point.
In broad discussions of carbon sequestration
strategies, this process is typically referred to as
carbon "saturation."4
The time it takes to reach this steady state varies
across soil types, site conditions, and management
practices. A key determinant of saturation time
is the land-use history of a given parcel—when
anthropogenic and natural disturbances occurred,
what land-use practices were involved, and how
long they persisted. If soils in the northern Corn
Belt, for example, were first tilled from native
grasslands with a given soil organic matter (SOM)
content in the early 20th century, cropped using
conventional tillage practices, and then converted
to lower-tillage practices, this land-use history
will strongly influence the level of SOM in the soils
today. Further, alternative management of these
soils to enhance SOM levels will be limited by the
difference between the current SOM level and the
potential or original level (see Figure 2-2).
Studies of soil conservation tillage effects on
carbon sequestration range from relatively quick
adjustment to steady state (e.g., 15 to 20 years
[West and Post 2002] to longer saturation periods
in excess of 50 years [Lai et al. 1998]; see Table 2-1).
The West and Post (2002) analysis reviews studies
of SOM changes from tillage and concludes that, in
most cases, saturation is reached at about 15 years,
with some residual carbon uptake after that period.
Figure 2-3 summarizes their analysis. Based on
their work, the analyses presented later in this
report use a soil saturation assumption of 15 years.
Forest carbon sequestration tends to saturate over
longer periods of time, 80 years or more after stand
establishment in the United States, varying by
Figure 2 2:
LD
Conceptual Model of Soil Organic Matter Decomposition and Accumulation Following
Disturbance
1
New steady-
state
IV
Time
Note: At steady state (II), carbon (C)
inputs from litter (L) equal C losses via
decomposition (D) (i.e., L/D = 1). After a
disturbance, D often exceeds L,
resulting in loss of C (II), until a new,
lower steady state is reached (III).
Adoption of new management, where L
exceeds D results in a reaccumulation
of C (IV) until a new, higher steady state
is reached (V). The eventual steady state
(A, B, or C) depends on the new
management adopted.
Source: Figure 4-5 in Kauppi and Sedjo
(2001), drawn from work of Johnson
(1995) and IPCC(2000).
4 It is necessary to make a scientific distinction between saturation, which refers to the ultimate biophysical limits to growth of an
ecosystem, and equilibrium, which refers to a system in steady state where inputs equal outputs. The latter is a subset of the
former. In other words, some systems can be in equilibrium, but not be at their biophysical saturation point, but if a system is at
its saturation point, it is also in equilibrium. By and large, our discussion of sequestration dynamics refers to the time it takes for
a system to reach its new equilibrium point after a land-use or land management change. In some cases, this new equilibrium
will not reflect the ultimate biophysical saturation point. However, to maintain consistency with typical word choice, we use the
term "saturation" to reflect the broad process of reaching a new equilibrium. For further discussion on the issue of soil carbon
saturation, see West and Six (2005).
2-10
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
Figure 2 3:
XX
\
\
Absolute Change in the Annual Rate of Carbon Sequestered Following a Change from
Conventional Tillage (CT) to No Till (NT)
Note: Estimates are relative to soil carbon values
under CT over the experiment duration, which
means the estimated change in annual sequestra-
tion is greater if carbon under CT is declining while
carbon under NT is increasing. Values in the figure
are absolute (no negative values) and represent the
percentage change in the estimated annual
sequestration rate, not the percentage change in
soil carbon. The method for calculating this value is
outlined by West and Post (2002). A nonlinear
regression curve has been fitted to the data, as
described by West et al. (2004), to indicate the
estimated peak and duration of soil carbon
sequestration. This estimate represents the
potential to sequester carbon, and soils or environ-
ments that have limiting factors that decrease or
inhibit soil carbon sequestration are represented by
values below the curve. Values considered as
statistical outliers are not shown in the figure.
"V
\
5 10 15 20
Experiment Duration (year)
25
Source: West and Post (2002).
forest type and site class (Birdsey 1996). Figure 2-4
illustrates a typical carbon growth pattern follow-
ing conversion of agricultural lands to a pine
plantation in the U.S. South. However, research
has shown that old growth forests in the United
States (e.g., Douglas fir or redwood stands in the
Pacific Northwest Westside [Harmon et al. 1990]
and in the tropics) may continue to accumulate
carbon for hundreds of years, although at a
decreased rate (Lugo and Brown 1986, Phillips
et al. 1998, Phillips et al. 2002a, 2002b).
Saturation has important implications for assess-
ing forestry and agricultural sequestration in the
United States, as saturation rates vary across carbon
pools, activities and land conditions. In the long
run, though, the rate at which activities accumulate
carbon at certain periods of time is not as critical to
climate change mitigation as the maximum, cumu-
lative carbon storage potential of the alternative
land use. Saturation is a dynamic phenomenon as
well and may respond to climate and/or future
environmental and technological change.
Reversibility of Carbon Sequestration
The accumulated carbon from forestry and agri-
cultural sequestration practices can be re-released
back to the atmosphere through either natural or
Figure 2 4: Carbon Accumulation on an Afforested Stand to Saturation
400
™" 350-
O1 300-
O
e. 250-
° 200-
3 150-
S 100-
* 50-
./^
/
/
/
/
^
Notes: 1) Saturation reached in about
year 80, and no additional carbon
sequest
50 t CO
Source:
0 20 40 60 80 100 120
Time (years)
ration afterward. 2) Soils contain
2 of soil organic matter in year 0.
Birdsey (1996).
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
2-11
-------
CHAPTER 2 • GREENHOUSE GAS MITIGATION OPTIONS IN U.S. FORESTRY AND AGRICULTURE
intentional disturbances, such as fires, manage-
ment changes, or logging. The climate benefits
of carbon sequestration activities are therefore
potentially reversible. This is sometimes referred
to as the permanence or duration issue. Note that
even if incentives for carbon sequestration, such as
those evaluated later in this report, cause harvests
to be delayed, harvesting may still occur eventu-
ally unless expressly prohibited by the incentive
program or policy.
Designing approaches for carbon sequestration
activities that appropriately capture the property
rights for the sequestered carbon and the liabilities
for carbon reversal remains a challenge. These
issues are examined further as part of the discus-
sions of Chapter 6.
Accounting for Carbon after Timber
Harvests
When timber is harvested, some of the carbon that
has accumulated over the years is removed from
the site and the rest is left on-site to decay over
time. The carbon that is removed from the site will
at any time following the harvest be in one of the
following carbon pools:
• products in use (very short-lived for paper, quite
long for lumber);
• landfilled, often stored for extended periods; or
• atmosphere through combustion (sometimes to
produce energy) or product decay.
Figure 2-5 illustrates the carbon flows over time
under rotational forestry. In addition to the carbon
fate after harvest discussed above, the figure
shows the reaccumulation of forest carbon in on-
site pools (trees, litter, soil) as a result of planting
trees after each harvest. The figure illustrates that
rotation forestry can continue to sequester carbon
over extended periods of time through the contin-
ued accumulation of carbon stored in products and
landfills. A complete accounting system should
capture all of these product flows.
Addressing Carbon Sequestration
Dynamics in this Report
In analyses presented later in this report, the
dynamics of saturation, reversibility, and post-
harvest destination of sequestered carbon are
handled within the framework of the FASOMGHG
model. As described in detail in Chapter 3, this
model comprehensively accounts for both carbon
sequestration and losses (i.e., sinks and sources)
in forestry and agriculture over time, including
harvested product pools. The accounting of both
carbon sinks and sources occurs in the baseline
and mitigation scenarios. Specific arrangements
for addressing reversibility risk are discussed in
Chapter 6.
Figure 2 5: Cumulative Carbon Changes for a Scenario Involving Afforestation and Harvest
o
n Emissions
n Energy
• Landfills
n Products
• Trees
• Litter
n Soils
0 10 20
Data Source: Birdsey (1996).
30 40 50 60 70 80 90 100 110 120
Time (Years)
2-12
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 3
Modeling Framework
and Baseline
Chapter 3 Summary
The FASOMGHG model is used to evaluate the joint economic and biophysical effects of GHG
mitigation scenarios in U.S. forestry and agriculture. This model includes all major GHG mitigation
options in U.S. forestry and agriculture and accounts for changes in CO2, CH4, and N2O, including
carbon sequestration and emissions over time. The model also generates estimates of nutrient
loadings and soil erosion in agriculture. FASOMGHG covers private timberlands and all agricultural
activity across the conterminous ("lower 48") United States, broken into 11 regions, and tracks
five forest product categories and more than 2,000 production possibilities for field crops, livestock,
and biofuels. FASOMGHG runs simulations for 100-year periods and reports results on a decadal
basis. The model simulates the actions of producers and consumers with perfect foresight of future
demands, yields, technologies, and GHG prices.
Mitigation analyses presented later in this report pivot off a FASOMGHG baseline (business as
usual) projection of future economic and GHG effects. This baseline estimates that private forests
will constitute a net carbon sink for several decades, though the sink is projected to diminish over
time. Direct (including N2O and CH4) and indirect sources and sinks in the forest and agriculture
sectors constitute a net emission source in the baseline of 270 Tg CO2 per year in the 2010 decade.
This net baseline emission rate nearly doubles by around 2030 and then stabilizes somewhat
thereafter. This pattern is largely dictated by carbon sink dynamics.
This chapter first presents the modeling
framework and data employed by the
FASOMGHG model of the U.S. forest and
agriculture sectors, which is the analytical founda-
tion for this report. After describing model details,
the chapter moves to the FASOMGHG business-
as-usual (BAU) baseline, focusing on future
projections of GHG emissions and sequestration
in the U.S. forest and agriculture sectors. The
FASOMGHG baseline is evaluated against recent
trends in land use, GHG emissions and sequestra-
tion, and baseline projections developed by other
recent studies.
Modeling Framework
Examining the dynamic role of forest and agricul-
tural GHG mitigation requires an analytical
framework that can depict the time path and GHG
consequences of forestry and agricultural activity.
To credibly model or simulate baseline and addi-
tional mitigation effects in these sectors, it is
critical to have as complete coverage as possible
along several key dimensions:
Sectoral
• Sufficient detail to identify targeted economic
opportunities within and across the sectors
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-1
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
(e.g., land-use change, forest management,
agricultural management, biofuel production).
• Inclusion of market-clearing processes and
resource competition needed to show the
commodity market (forest and agricultural
products) feedback effects of mitigating GHGs
in forestry and agriculture.
Spatial
• Heterogeneity of biophysical and economic
conditions within and across regions as it relates
to the production of food, fiber, fuel, and the
GHG consequences thereof. For instance,
regional carbon sequestration rates can vary
spatially by more than an order of magnitude.
• Competition for region-specific resources, such
as land and water, which affects economic
responsiveness in forestry and agriculture to
traditional commodity market signals and to
GHG economic incentives.
Temporal
• Ability to capture dynamic biophysical process-
es (e.g., soil and biomass carbon accumulation
over time, fate of harvested wood products).
• Ability to capture dynamic economic processes
(investment, technological progress, demand
trends, traditional commodity, and GHG market
developments).
In addition, models used for policy evaluation
should, to the extent possible, be calibrated to and
validated by observed economic and biophysical
phenomena. FASOMGHG encompasses the
dimensions just defined and thereby provides an
analytical foundation to address the issues raised
in this report. This section of the report describes
FASOMGHG's conceptual framework, scope of
coverage, data, and other details.
General Model Description
FASOMGHG is an augmented version of the
Forest and Agricultural Sector Optimization Model
(FASOM) (Adams et al. 1996) as developed by
Lee (2002). The model has all of the forest- and
1 For more complete model detail on FASOMGHG and its affiliated models, consult Dr. Bruce McCarl's Web site,
(http://agecon2.tamu.edu/people/faculty/mccarl-bruce/papers.htm).
agriculture-sector economic coverage of the
original FASOM model unified with a detailed
representation of the possible mitigation strategies
in the agriculture sector adapted from Schneider
(2000) and McCarl and Schneider (2001).
FASOMGHG is a 100-year inter temp oral, price-
endogenous, mathematical programming model
depicting land transfers and other resource
allocations between and within the forest and
agriculture sectors in the United States. The model
solution portrays a multiperiod equilibrium on a
decadal basis. The results from FASOMGHG yield
a dynamic simulation of prices, production,
management, consumption, and GHG effects
within these two sectors under the scenario
depicted in the model data.
FASOMGHG can simulate responses in the U.S.
forest and agriculture sectors to economic incen-
tives such as GHG prices or mitigation quantity
targets. Economic responses include changes in
land use between and within the sectors and
intrasectoral changes in forest and agricultural
management.
FASOMGHG's key endogenous variables include
• land use;
• management strategy adoption;
• resource use;
• commodity and factor prices;
• production and export and import quantities;
and
• environmental impact indicators:
— GHG emission/absorption (CO2, CH4, N2O)
and
— surface, subsurface, and groundwater pollu-
tion for nitrogen, phosphorous, and soil
erosion.
Table 3-1 summarizes FASOMGHG's key dimen-
sions. The remainder of the section provides more
detail on the model's structure, data, and key
parameters.1
3-2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Table 3-1: FASOMGHG Model: Key Dimensions
lodel Dimension Forest Sector
General scope and coverage
Geographic coverage
Regional detail
Land ownership coverage
Economic dimensions
Economic modeling approach
Time horizon
Discount rate
Commodities
Price and cost data
Supply/land inventory
Supply/biophysical yield
Demand
International trade
Environmental variables
GHG coverage
Non-GHG environmental
indicators
Land coverage for conterminous United
States with other regions linked by
international trade
11 U.S. regions, 9 of which produce
forest goods
All private timberland in conterminous
United States
Optimizing producer and consumer
behavior over finite time horizon
Model base year = 2000
Resolution = 10-year time steps
Typically run for 100 years
4%
10 commodities
5 products: sawlogs, pulpwood,
fuelwood and milling residues (2)
2 species: softwood and hardwood
Resource Planning Act (RPA)
assessment (USDA Forest Service 2003)
USDA Forest Service Forest Inventory
and Analysis Data
USDA Forest Service ATLAS model
(Mills and Kincaid1992)
Adapted from demand models used in
latest RPA Assessment (USDA Forest
Service 2003)
10 excess-demand regions facing each
timber-producing region plus Canada
CO2 as carbon sequestration in forest
ecosystem pools and in harvested
wood products
Timberland area by region, species,
owner, age class
Forest management intensity
Same
11 U.S. regions, 10 of which
produce agricultural goods
All agricultural land in major
commodity production in the
conterminous United States
Same
Same
Same
48 primary products
45 secondary products
USDA NRCS data with updates
based on Agricultural Statistics
USDA NRI, Agricultural Census,
and MASS data
Crop budgets and EPIC (Williams
et al. 1989) model simulations
Variety of demand studies (see
"Agricultural Product Demand"
on page 3-9)
28 international regions for the
main traded commodities plus
excess supply and demand for
others
CO2 sequestration and emissions
CH4 emissions
N2O emissions
Agricultural land allocation
Tillage practices
Irrigation water use
Cropland loadings of nitrogen,
phosphorous, potassium, erosion,
and pesticides
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-3
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Geographic Coverage/Regional Detail
FASOMGHG covers forest and agricultural activ-
ity across the conterminous ("lower 48") United
States, broken into 11 separate regions (see Table
3-2 and Figure 3-1).
The 11 regions are a consolidation of regional
definitions that would otherwise differ if the forest
and agriculture sectors were treated separately.
The forest sector considers nine major production
regions and agriculture distinguishes 10 regions.2
The 11-region breakdown reflects the existence
of regions for which there is agricultural activity
but no forestry, and vice versa. For instance, the
Northern Plains (NP) and Southwest (SW) regions
reflect important differences in agricultural
characteristics, but no forestry activity is included
in either region. Likewise, there are important
differences in the two Pacific Northwest regions
(PNWW, PNWE) for forestry, but only the PNWE
region is considered a significant producer of the
agricultural commodities tracked in the model.
Land Base
FASOMGHG covers all cropland and pastureland
in production throughout the conterminous United
States. Livestock grazing is also tied to the use of
animal unit months (AUMs) on public rangelands,
largely in the western states. The model accounts
for timber production from all U.S. forestlands,
private and public, and timber imports. However,
the forest-sector mitigation activities and GHG
(carbon) accounting are limited to private timber-
land in the conterminous United States. Mitigation
and carbon flows from public timberland and all
forestlands too unproductive to be considered
timberland are excluded from the model because
of data limitations and because model development
has heretofore focused on potential mitigation
responses of the private sector to market-based
incentives.3 The potential impact of excluding
public lands from the forest-sector analysis is
addressed further below.
General Economic Concepts: Optimizing
Behavior
At its heart, FASOMGHG solves a constrained
dynamic optimization problem defined as follows:
Objective Function: Maximize the net present
value (NPV) of the sum of producer and consumer
surpluses across the forest and agriculture sectors
over time (100 yrs), including any GHG payments
introduced by a mitigation scenario.
Constraints:
• Total production = total consumption
• Technical input/output relationships hold
• Land-use balances
By maximizing the sum of producer and consumer
surplus, the model ensures that all suppliers and
demanders are making optimal choices about what
to produce and consume. Because these choices
occur over time, the optimizing nature of the
model assumes that producers and consumers
have perfect foresight regarding future demands,
yields, technologies, and prices. See Box 3-1.
Given that the model is defined for a finite period,
there will be immature trees of some age at the
end. If the model did not place a value on these
forests, the optimizing nature of the model would
be inclined to deplete all timber at the end of the
projection period rather than leave it around for
future harvests. Similarly, agricultural land values
at the end of the period must also be considered to
ensure that land is not inappropriately converted
as a result of a perceived lack of opportunity cost.
To counter these ending-period anomalies, terminal
conditions are imposed on the model that value
ending immature trees and land remaining in
agriculture. FASOMGHG assumes that forest
management is, from the last period onward, a
continuous or constant flow process with a forest
inventory that is "fully regulated" on rotations
equivalent to those observed in the last decades
2 The 10 agricultural regions in FASOMGHG are an aggregation of the 63 agricultural regions considered in the agriculture-only
version of this model (ASMGHG) (Schneider 2000).
3 Timberland is all land with forest cover capable of generating at least 20 cubic feet per acre per year of merchantable timber.
Land with forest cover that does not meet this criterion is considered unproductive forestland.
3-4
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Table 3-2: FASOMGHG Regional Definitions
CB
NP
LS
NE
PNWE
PNWW
PSW
RM
SC
SE
SW
Corn Belt
Northern Plains
Lake States
Northeast
Pacific Northwest-east side
Pacific Northwest-west side
Pacific Southwest
Rocky Mountains
South-Central
Southeast
Southwest
Illinois, Indiana, Iowa, Missouri, Ohio
Kansas, Nebraska, North Dakota, South Dakota
Michigan, Minnesota, Wisconsin
Connecticut, Delaware, Maine, Maryland, Massachusetts,
New Hampshire, New Jersey, New York, Pennsylvania,
Rhode Island, Vermont, West Virginia
Oregon and Washington, east of the Cascade mountain range
Oregon and Washington, west of the Cascade mountain range
California
Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Utah,
Wyoming
Alabama, Mississippi, Louisiana, Eastern Texas, Eastern
Oklahoma, Arkansas, Tennessee, Kentucky
Virginia, North Carolina, South Carolina, Georgia, Florida
Western Texas, Western Oklahoma
Figure 3 1: FASOMGHG Regions
Pacific Northwest
- East side
Pacific Northwest
- West side
Northeast
Northern Plains
(agriculture only)
Pacific
Southwest
Southeast
Southern Plains
(agriculture only)
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-5
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Box 3-1: Perfect Foresight in Climate
Economic Models
Three main approaches to economic modeling of
climate change mitigation have been used in the
past 2 decades. Engineering cost curves use activity
data and cost data to compare and order mitigation
practices of technologies by region from lowest to
highest cost. Econometric approaches use revealed
preferences of landowners for activity and cost data
but do not include feedbacks in the land and com-
modity markets over time. Most climate economic
models of multiple sectors, including FASOMGHG,
use the third approach, dynamic simulation, which
explicitly models economic decisions and market
outcomes over time subject to an underlying
behavioral or process model.
Weyant (2000) identifies foresight as a key element
of structure for dynamic climate economic models,
with two prevailing options: perfect and myopic.
FASOMGHG employs the perfect foresight option,
as do all but one of the climate economic models
reviewed by Weyant. Perfect foresight assumes that
agents, when making decisions that allocate resourc-
es over time (e.g., investments), know with certainty
the consequences of those actions in present and
future time periods.
Landowners understand that decisions they make
today, such as converting agricultural land to trees,
depend on their expectations of future prices and
yields in forestry and agriculture and, in this case,
prices and yields of GHGs. FASOMGHG simulates
these decisions and employs these predictions to
determine which actions should be taken today and
which deferred to the future. As Weyant points out,
this form of perfect foresight allows for an efficient
allocation of resources over time. These perfect
foresight models are also classified as dynamic
optimization models. In contrast, myopic foresight
uses no predictions of future prices and yields and
uses only current information to make decisions that
affect resource allocation over time, although not as
efficiently as under perfect foresight.
In reality, investors have neither perfect foresight nor
perfect myopia, so the modeling decision is not about
which assumption is factually correct. In practice,
perfect foresight is the approach preferred by most of
the climate economic modeling community because
of its consistency with economic theory and efficiency.
But it is important to understand the implications of
the modeling decision. In short, the costs of GHG
mitigation estimated using perfect foresight models
such as FASOMGHG will tend to reflect a more
efficient mitigation response and thus be lower than
costs estimated using a myopic foresight model.
of the projection (see Adams et al. [1996]). The
terminal value of land remaining in agriculture is
formed by assuming that the last period persists
forever.
The multiperiod nature of the economic problem
requires transforming future revenues and costs to
the present using a real (inflation-adjusted) annual
discount rate. The default rate used in FASOMGHG
is 4 percent, which is broadly consistent with oppor-
tunity costs of capital in agriculture and forestry.
Forest-Sector Economic Detail
The forest-sector component of FASOMGHG is
derived from the USDA Forest Service modeling
system for performing periodic assessments of the
nation's forests and related renewable resources
under the Resources Planning Act (RPA). For more
information on the RPA timber market modeling
framework, see USDA Forest Service (2003).
Forest Commodities
FASOMGHG tracks the following five forest
product categories:
• logs (3): sawtimber, pulpwood, fuelwood
• residues (2): logging and milling
These products are differentiated by two species
types (softwood and hardwood) for a total of 10
forest commodities.
Forest Product Supply
Log supply in the model is based on a "model II"
even-aged harvest scheduling structure (Johnson
and Scheurman 1977) allowing multiple harvest
age possibilities. The model's forest inventory is
tracked by age, and the harvest responses are
limited to even-aged management, wherein a forest
stand is grown to a certain age and then harvested
and regrown (unless land is allocated to another use
after harvest). Timber harvests are responsive to
the market price, discount rate, and growth rate of
the forest stand. Log supply is volume harvested in
each period, so endogenous decisions at the forest
level are
4 The forest production regions include 9 of the 11 regions identified in Table 3-2. The omitted regions are the Northern Plains and
Southwest, which do not include any appreciable timber production.
3-6
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
• length of rotation,
• management regime to regenerate the harvest-
ed area, and
• species for regeneration.
Supply is segmented into two private-sector
classes (industry and nonindustrial private) and
nine regions within the United States.4 Harvests
from public lands are included in the model but
are exogenously determined, rather than solved
by the model.
Timber supply comes from harvests of the mer-
chantable timber inventory existing at that time.
The model's timber inventory data are derived
from USDA Forest Service Forest Inventory and
Analysis (FIA) field data. FIA is essentially a survey
of U.S. forests, drawing data from approximately
70,000 field plots nationwide. These field plots
have been sampled over time since the 1930s, with
survey timing varying by region. The version of the
FASOMGHG model used in this report is based on
FIA data from the early 1990s.5 The timber inven-
tory is stratified by the following dimensions:
• region (9),
• land class defining suitability for movement
between forestry and agriculture (5),
• ownership (2),
• forest type (4),
• site productivity class (3),
• timber management intensity class (4), and
• 10-year age classes (10).
For timber supply modeling purposes, the critical
biophysical element of the timber inventory is the
merchantable yield volumes. These volumes are
tracked in the inventory data, and FASOMGHG
models their evolution over time using the ATLAS
model (Mills and Kincaid 1992), which essentially
keeps inventory balances over time by tracking for
each stratum in the inventory its volume growth,
volume harvested, old area out, and new area in.
Each stratum is represented by the number of
CHAPTERS • MODELING FRAMEWORK AND BASELINE
timberland acres and the growing stock volume
per unit area.
Forest Product Demand
The 10 forest commodities listed above are the raw
materials produced by the forest sector that are
ultimately used in the production of final products
used by consumers. Therefore, forest commodity
demand is characterized as a derived demand for
these commodity inputs to the sector's final prod-
ucts. Final product demand is based on the Timber
Assessment Market Model (TAMM) (Adams and
Haynes 1996) for solid wood products and the
North American Pulp and Paper (NAPAP) model
(Zhang et al. 1996) for pulp and paper products.
The derived demand system starts with the
demand for final products, which include the
broad categories of lumber, plywood, oriented
strand board (OSB), paper, paperboard, and
market pulp, and the demand for wood as a fuel.
Final product demand is converted to raw material
demand (logs and residues) via physical conver-
sion factors. Substitution is allowed between raw
materials in a downward hierarchy from sawlogs
to pulpwood to fuelwood, meaning that sawlogs
can be used in lieu of pulpwood in pulp and paper
production, but not vice versa. Likewise, pulpwood
can be used in lieu of fuelwood, but not vice versa.
Additionally, mill residues from sawlog processing
can be used as a raw material to pulp and paper
production. Total raw material demand is bound
by sector processing constraints, which is also
endogenous to the model.
The product demand functions shift over time as
a function of
• macroeconomic factors (gross domestic product
[GDP], population, labor force) and
• other key structural shifts:
— housing starts,
— pulp and paper technical factors (e.g.,
recycling), and
— log conversion factors.
5 The model is currently being updated to reflect data from the early 2000s.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-7
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
model assumes continuation of the current trade
policy environment.6
Agriculture-Sector Economic Detail
The agriculture-sector component of FASOMGHG
is derived from two predecessors, the Agricultural
Sector Model (ASM) (Chang et al. 1992) and
ASMGHG (Schneider 2000), both of which are
static models of the U.S. agriculture sector. For
consistency with the time dynamics introduced
by the forest sector, economic decisions in the
agriculture sector also conform to the intertemporal
welfare maximization approach described above.
Agricultural activity within each decade is assumed
constant, with dynamic updating each decade
based on USDA Economic Research Service (ERS)
projections of future yield and consumption trends
and past consumption and production trends,
where available.
The macroeconomic and other structural shifts in
demand are based on 50-year projections devel-
oped for the USDA Renewable Resource Planning
Act Assessment and described in its supporting
documentation (USDA Forest Service 2003).
International Trade in Forest Products
Canada is the dominant forest products trading
partner with the United States, with Canadian
exports accounting for a sizable share of total U.S.
consumption of softwood lumber and some pulp
and paper products, such as newsprint. Therefore,
Canada-U.S. final product trade flows are treated
explicitly in the model. Exports/imports from
countries other than Canada are aggregated as
price-sensitive net trade functions facing the U.S.
regional markets. Future trade is projected to shift
in response to exchange rate projections. The
Table 3-3: Agriculture-Sector Commodities
Primary Products
• Crops: Cotton, corn, soybeans, soft white wheat, hard red winter wheat, Durham wheat, hard red spring wheat,
sorghum, rice, oats, barley, silage, hay, sugarcane, sugarbeets, potatoes, tomatoes for fresh market, tomatoes
for processing, oranges for fresh market, oranges for processing, grapefruit for fresh market, grapefruit for
processing, rye
• Animal products: Grass-fed beef for slaughter, grain-fed beef for slaughter, beef yearlings, calf for slaughter, cull
beef cows, milk, cull dairy cows, hogs for slaughter, feeder pigs, cull sows, lambs for slaughter, lambs for
feeding, cull ewes, wool, unshorn lambs, mature sheep, steer calves, heifer calves, vealers, dairy calves, beef
heifer yearlings, beef steer yearlings, dairy steer yearlings, heifer yearlings, other livestock, eggs, broilers, turkeys
• Biofuels: Willow, poplar, switchgrass
Secondary Products
• Crop related: Orange juice, grapefruit juice, soybean meal, soybean oil, high fructose corn syrup, sweetened
beverages, sweetened confectionaries, sweetened baked goods, sweetened canned goods, refined sugar, gluten
feed, starch, refined sugar cane, corn oil, corn syrup, dextrose, frozen potatoes, dried potatoes, chipped pota-
toes
• Livestock related: Fluid milk, grain-fed beef, grass-fed beef, veal, pork, butter, American cheese, other cheese,
evaporated condensed milk, ice cream, nonfat dry milk, cottage cheese, skim milk, cream, chicken, turkey
• Mixed feeds: Cattle grain mix 0, cattle grain mix 1, high-protein cattle feed, broiler grain, broiler protein, cow
grain, cow high protein, range cubes, egg grain, egg protein, pig grain, feeder pig grain, feeder pig protein, pig
farrowing grain 0, pig farrowing grain 1, pig farrowing protein, pig finishing grain, pig finishing grain 1, pig
finishing protein, dairy concentrate, sheep grain, sheep protein, stocker protein, turkey grain, turkey protein
• Biofuels: MMBtu of power plant input, ethanol, market gasoline blend, substitute gasoline blend
6 For more on forest-sector trade and demand projection assumptions used in FASOMGHG, see USDA Forest Service (2003),
Chapter 2.
3-8
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Agricultural Commodities
The model's agriculture sector encompasses both
primary production and secondary processing/
conversion, as indicated in Table 3-3.
Agricultural Product Supply
Primary commodity production is derived from
allocation decisions based on a set of more than
2,000 production possibilities for field crops,
livestock, and biofuels. The allocation decisions
are based on optimizing across the budgets
associated with each production possibility, given
prices for outputs and inputs. Budgets are based
on using inputs to produce a given level of outputs.
Land is available in five cropland categories (based
on erodibility) plus pastureland. The use of erod-
ibility to classify cropland enables estimation of
soil erosion and other environmental effects from
different cropping and management practices, as
reported in Chapter 7. The land inventory is fixed
but can migrate back and forth between agricul-
ture and forestry. Inputs are either regionally
supplied subject to a price-sensitive input supply
function (labor, grazing, and irrigation water) or
nationally supplied at a fixed price (energy, agri-
cultural chemicals, and equipment in more than
100 categories). Supply emanates from 10 regions
within the United States.7
In the first 2 decades, the production solution is
required to be within the combination of crop
mixes observed historically, following a method
developed by McCarl (1982), but is free to vary
thereafter. Agricultural yields and factor usage
vary by decade with USDA ERS-projected and
historical trends in yield growth and input
requirements to sustain this yield growth based
on Chang et al. (1992).
Primary commodities are converted to secondary
products via processing activities with associated
costs (e.g., soybean crushing to meal and oil,
livestock to meat and dairy). Processed products
and some primary commodities are supplied to
meet national-level demands. Once commodities
are supplied to the market, they can go to livestock
use, feed mixing, processing, domestic consump-
tion, or export.
Agricultural Product Demand
The model uses constant demand elasticity func-
tions to represent domestic and export demand.
International agricultural demand is adapted from
the USDA SWOPSIM model (Roningen et al. 1991).
Domestic demand is drawn from many studies
plus computations of arc elasticities from various
other sources (Baumes 1978, Burton 1982, Tanyeri-
Abur 1990, Schneider 2000, Hamilton 1985).
Product demands are updated each decade based
on USDA ERS projections and on historic trends
where USDA data are unavailable.
International Trade in Agricultural Products
FASOMGHG has explicit trade functions between
the United States and 28 distinct foreign trading
partners for agricultural commodities having such
detailed trade data available. For the remaining
commodities traded internationally, excess supply/
demand functions are specified to capture net
trade flows with the rest of the world as one
composite trade region with the United States.
Demand levels are parameterized based on
SWOPSIM and USDA annual statistics.
Biofuels
For the purposes of this analysis, biofuels are
treated as another agricultural commodity, but as
shown in subsequent chapters of the report, they
have a rather large potential for GHG mitigation
within the sector and thereby warrant special
attention. The data used in the analysis for bio-
mass production conditions were mainly obtained
from Oak Ridge National Laboratory (ORNL).
The data from ORNL include average yields for
the three biomass crops (willow, switchgrass, and
hybrid poplar) and their corresponding farm-level
production costs, varying by state. Estimates of
hauling costs are added to the farm-level produc-
tion costs to complete the budget data needed for
the production model.
7 The agricultural production regions match 10 of the 11 regions identified in Table 3-2. The omitted region is the Pacific North-
west-west side.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-9
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
On the demand side, special consideration was
given to the possibility that infrastructure limita-
tions in the energy sector might impede rapid
increase in market penetration for biofuel crops,
given the very low use of biofuel crops to date.
Therefore, market penetration constraints were
imposed on biofuel demand for each decade in the
model, with the initial constraints being relaxed
over time as more capacity develops. These con-
straints were developed in consultation with staff
from the U.S. Department of Energy's (DOE's) EIA,
drawing on work from Haq (2002).8
Cross-Sector Land Interaction
A defining element of FASOMGHG is its ability
to allocate land across and within the forest and
agriculture sectors in response to economic and
biophysical forces. As shown in Figure 3-2, the
model includes four primary choices of land
transfers: from forest to agriculture (cropland or
pastureland), agriculture (cropland or pasture-
land) to forestry, cropland to pasture, and pasture
Figure 3 2: FASOMGHG Market Linkages
Forest-Sector Model
(TAMM Based)
Timberland
- Public
— Forest industry
— Nonindustrial private
Nonconvertible forest
Convertible forestland
— Region
— Soft and hard
— Prod, class
— Mgt. class
to cropland. Many forested tracts are not suitable
for agriculture because of topography, climate, soil
quality, or other factors, so the model accounts for
land that is not mobile between uses and land that
is. Costs for converting forestland reflect differ-
ences in site preparation costs because of stump
removal amounts, land grading, and other factors.
Greenhouse Gas Accounting
Table 3-4 lists the GHG sinks and sources covered
by FASOMGHG by sector and gas.
Forest-Sector GHG Accounting
Forest ecosystem carbon accumulates in the forest
in four distinct pools: trees, understory vegetation,
litter, and soils. The allocation of carbon among
these components varies by region, forest type,
stand age, site quality, and previous land use.
Within FASOMGHG, these allocations are derived
from the USDA Forest Service FORCARB model
(Birdsey 1992) and Turner et al. (1993). Critical
among these relationships is the role of time.
CROPFOR
FORCROP
PASTFOR
FORPAST
Agricultural-
Sector Model
Agricultural Land
— Ag-only land
Convertible cropland
Convertible
pastureland
Urban, Developed and Special Uses
8 For more complete model detail on FASOMGHG and its affiliated models, consult Dr. Bruce McCarl's Web site, (http://agecon2.
tamu.edu/people/faculty/mccarl-bruce/papers.htm).
3-10
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
As described in Chapter 2, once a forest is estab-
lished, it typically accumulates carbon steadily for
several decades, then the sequestration rate begins
to decline. If the forest is left in place without
harvest or other disturbance, the growth rate may
eventually diminish when the forest reaches a
steady-state equilibrium.9 The carbon accounting
component of FASOMGHG captures these nonlin-
ear biophysical growth effects.
Additionally, forest carbon accumulates in harvested
wood products after it leaves the forest. The carbon
can reside in the products while they are being
used (e.g., lumber and plywood in housing) or
in landfills after the products are discarded and
before they decompose and are re-released to
the atmosphere. Storage in wood produces can
continue for a very long time after harvest. The
parameters used to allocate the wood product
carbon destination over time after harvest are
derived by the HARVCARB model (Row and
Phelps 1991).
After it is harvested, carbon can be burned in
the production process and released back to the
atmosphere. If the burning occurs as part of a
combustion process to generate bio-energy, the
releases can be viewed as a form of fossil fuel
substitution. This form of substitution could be
accounted for differently than a normal emission
release because it foregoes the transfer of below-
ground carbon (coal, petroleum, gas) to the atmo-
sphere, replacing it with "recycled" biofuel.
Therefore, FASOMGHG tracks the amount of
forest carbon burned for biofuel to examine policy
scenarios under which this carbon is treated
separately.
The combination of carbon accumulation in forest
ecosystems, harvests, releases, product storage,
and biofuel energy offsets can create an interest-
ing carbon dynamic over time from the forest
sector, as shown in Figure 3-3.
Agriculture-Sector GHG Accounting
As with forests, carbon accumulates in agro-
ecosystems; although in the case of U.S. agriculture,
sequestration occurs largely in the form of soil
organic carbon (SOC), rather than biomass.
FASOMGHG captures SOC changes in response
to cropping patterns and tillage changes, based on
the CENTURY model (Parton 1996). Three types
of tillage are depicted: conventional, minimum
tillage, and zero tillage. Four different fertilization
Table 3-4: GHG Emission Sources and Sinks in FASOMGHG
CO2 Sinks/Sources
Sector (biomass and soil carbon)
Fossil Fuel CO,
CH4 Sources
N2O Sources
Forest Carbon sequestration
and release from forest
ecosystems and harvested
wood products
Agriculture Carbon sequestration
and release from agro-
ecosystem soils
Biofuel use in wood
processing as a fossil
fuel emission offset
On-farm energy use
Energy associated with
inputs (e.g., fertilizer
production)
Biofuel production and
use as a fossil fuel
emission offset
Livestock manure Fertilizer use
Livestock enteric Residue burning
fermentation
Rice cultivation
Livestock manure
9 As explained in Chapter 2, this carbon steady state is sometimes referred to as a "saturation point," but equilibrium is a more
scientifically precise term. A site can be in steady state, with system inputs and outputs in balance and no net sequestration
taking place yet still be able to yield more carbon if, say, inputs were increased by natural (CO2 fertilization) or artificial
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-11
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Figure 3 3: Cumulative Carbon Changes for a Scenario Involving Afforestation and Harvest
HI
O 400
D Emissions
D Energy
• Landfills
D Products
• Trees
• Litter
D Soils
0 10 20 30 40 50 60 70 80 90 100 110 120
Time (years)
Data Source: Birdsey (1996).
levels are also modeled, and crops are simulated
by region. Soil carbon sequestration is assumed to
occur at a constant rate for 15 years and then
stabilizes thereafter, based on the work of West
and Post (2002). Land can move to less intensive
tillage with carbon gains or to more intensive
tillage with carbon losses.
The agriculture sector releases CO2 to the atmo-
sphere through the on-farm use of fossil fuels as an
energy source (tractors, irrigation, drying opera-
tions) and through the upstream emission of fossil
fuels in the production of other material inputs
such as agricultural chemicals using calculations
from Schneider (2000) based on USDA data.
The agriculture sector is a major source of the
non-CO2 gases—CH4 and N2O. CH4 releases in
agriculture are from enteric fermentation, manure
management, and rice cultivation. Enteric fermen-
tation emissions and emission changes from the
baseline are estimated using data based on EPA
data and a set of alternatives proposed by Johnson
et al. (2003a, 2003b), involving changes in feeding
regimes, improved pasture use, and use of bovine
somatrophine (bST). Manure emissions are
estimated using swine and dairy farm data esti-
mated for digester use based on EPA data. Rice
CH4 emission are estimated using data used to
support the U.S. national GHG inventory (EPA
2003). N2O sources in agriculture come from
fertilizer use, residue burning, and livestock
manure. These N2O releases are estimated using
U.S. activity data with IPCC emissions factors.
Difference in Scope of GHG Accounting
in the Forest and Agriculture Sectors
Forest-sector GHG accounting in FASOMGHG
does not include CO2 emissions from on-site
machinery and upstream processing of inputs,
CH4 emissions from forested wetlands or landfilled
forest products, nor N2O emissions from fertilizer
use. Most of emissions data for these activities or
sources are not readily available for the forest
sector. Thus, the GHG accounting for the forest
sector has a narrower scope than for the agricul-
ture sector in FASOMGHG. However, the omitted
emissions in the forest sector are generally thought
to be small relative to those included, so their
omission is unlikely to create a distorted view of
mitigation potential in this report.
3-12
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Non-GHG Environmental Indicators
Several variables discussed above provide useful
information on environmental quality implications
of modeled outcomes. In the forest sector, these
include forest land area composition by species
and age class, forest management intensity, and
rotation length (harvest age). Land-use and
management patterns are also reported on the
agriculture side of the model. In addition, the
model draws from the agricultural management
model EPIC (Williams et al. 1989) to produce data
on irrigated acres and water use and on cropland
loadings of nitrogen, phosphorous, potassium,
erosion, and pesticide use.
GHG Mitigation Strategies
The comprehensive coverage of FASOMGHG
allows for the identification of several basic strate-
gies for GHG mitigation in forestry and agricul-
ture. Table 3-5 lists broad mitigation strategies
aligned with specific mitigation activities tracked
by FASOMGHG. These strategies are a mix of
sequestration, emissions reduction, and fossil fuel
offsets. Although each strategy has a focal GHG
of interest, it is important to recognize that
FASOMGHG incorporates multi-GHG accounting
and therefore captures the net GHG consequences
of each strategy. This is particularly critical given
that GHG policies may include only a subset of
GHGs, as discussed further in Chapter 5.
While FASOMGHG is fairly complete in its
coverage of GHG mitigation opportunities in U.S.
forestry and agriculture, some mitigation opportu-
nities remain outside the scope of the model.
Of those activities referenced in Chapter 2, two
warrant further discussion here (see Table 3-6).
First, the model does not consider forest manage-
ment opportunities on the 275 million acres (37
percent) of all forestland in the United States in
public ownership (Smith et al. 2001). Assuming
all of those acres could be managed to achieve the
carbon enhancements for forest management
Table 3-5: Broad GHG Mitigation Strategies Covered in FASOMGHG
Strategy
Mitigation Activities Tracked in FASOMGHG
Target GHG
Afforestation
Convert agricultural lands to forest
CO,
Forest management
Lengthen timber harvest rotation
Increase forest management intensity
Forest preservation
Avoid deforestation
CO,
Agricultural soil carbon sequestration
Crop tillage change
Crop mix change
Crop fertilization change
Grassland conversion
CO,
Fossil fuel mitigation from crop
production
Crop tillage change
Crop mix change
Crop input change
Irrigated/dry land mix change
CO,
Agricultural CH4 and N2O mitigation
Crop tillage change
Crop mix change
Crop input change
Irrigated/dry land mix change
Enteric fermentation control
Livestock herd size change
Livestock system change
Manure management
Rice acreage change
CH4
N,O
Biofuel offsets
Produce crops for biofuel use
CO,
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-13
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Table 3-6: Mitigation Options Not Explicitly Captured in FASOMGHG
Option Description
Maximum Biophysical
Mitigation Potential
Economic and Other
Adoption Factors
Forest management
on public lands
Grazing land
management
Enhancing forest
carbon through
changes in
management of
publicly owned
forestlands
Improving forage
quantity and quality
to retain more soil
carbon
-685 Tg CO2 per year
(2 75 M M acres at 2. 5 t
CO2 per acre per year)
-590 Tg CO2 per year
(590 MM acres of
nonfederal pasture/
rangeland at 1 t CO2
per acre per year)
Public lands are by mandate
managed for multiple uses,
implying an opportunity cost
of managing specifically
for carbon. Allowable federal
timber harvest levels set by
Congress could have a large
impact on baseline levels of
carbon storage.
Limited data are available on
the cost of adopting practices and
corresponding carbon and other
GHG effects.
referenced in Chapter 2 (roughly 2.5 t CO2 per
acre per year), this could hypothetically enhance
forest carbon sequestration by nearly 700 Tg CO2
per year.
However, this maximum biophysical potential
estimate has little meaning. The biophysical
productivity of public forestlands is generally
lower than private lands, and this is an estimate
of pure biophysical potential, without considering
economic or other institutional factors. There is
no information on the costs of achieving this
mitigation on public forests. Moreover, the analy-
ses in this report gauge the response of the forest
and agriculture sectors to GHG prices or market
incentives, essentially a private-sector phenom-
enon. Public land responses are possible but
require public land management legislative
mandates (e.g., changes in national or state forest-
land harvest or planting levels) that are fundamen-
tally different from the market-based approaches
addressed in this report.
Another set of strategies not captured in
FASOMGHG is grazing land management prac-
tices. Grazing land includes rangeland, pasture-
land, and grazed forestland. The United States
has about 590 million acres of nonfederal grazing
land (USDA NRCS 2000). Little data exist on either
the carbon sequestration effects or costs of these
changes in practices. Using a mid-range estimate
of 11CO2 per acre per year for grazing practices
from Chapter 2, this suggests a maximum biophys-
ical potential for mitigation of nearly 600 Tg CO2
per year. But again, little data are available from
which to conduct economic analyses of these
options. In addition, changes in grazing practices
could be adopted on federal lands, but limited
information is available on the area of land to
which these practices could be applied, the cost,
and the consistency with other public land man-
agement objectives.
One other category of practices that is implicitly
captured in FASOMGHG but is not broken out
separately is riparian buffer establishment. As
indicated above, riparian buffers are the establish-
ment of vegetative cover such as grass or trees near
water bodies. The model captures afforestation
and grassland conversion, but it does not have the
data to determine whether those conversions are
taking place in riparian areas. Therefore, the
model will implicitly capture establishment
of trees and grasses in this area in response to
the GHG incentives put forth (e.g., GHG price
payments), but it will not be able to identify this
distinctly as riparian buffers. As a result, the
model cannot currently examine policies specifi-
cally aimed to increase riparian buffers.
3-14
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Baseline GHG Projections from the
Forest and Agriculture Sectors
The estimation of a baseline is an important first
analytic step for this study, because the analyses of
GHG mitigation potential presented in subsequent
chapters must be measured against a credible
baseline reflecting a continuation of BAU activity.
The analysis begins by using the FASOMGHG
model to simulate future economic activity and
corresponding GHG effects in the forest and
agriculture sectors under a continuation of the
status quo, or BAU. Departures from this baseline
constitute the mitigation quantities estimated in
response to the price and policy scenarios analyzed
throughout this report.
FASOMGHG Baseline Projections
This section presents baseline projections from the
FASOMGHG model. These results reflect model
outputs when FASOMGHG is run based on the
exogenous data and trends discussed above and
without any GHG policies in place. We look first at
projections of key land-use and management
trends and see how these comport with trends
reported in recent land-use inventory data. We
then look at the FASOMGHG projections of the
sectors' GHG flows (emissions and sequestrations)
and compare these projections with other second-
ary sources as well.
Baseline Land-Use and Management
Projections
One of the driving factors of the GHG effects in
these sectors is how land is expected to be used
over time. FASOMGHG simulates land allocation
for each region across time. National-level projec-
tion of land use across the major categories of
cropland, pasture/range, timberland, and devel-
oped use is illustrated in Figure 3-4. Cropland is
projected to decline steadily into the future as
productivity improvements reduce the demand
for cropland relative to other uses. This is a
continuation of recent history, as discussed below.
Pasture/range land is projected to rise over time,
as demand for livestock products is projected to
grow. Timberland is projected to decline just
modestly over time, as demand for timber attracts
some land from agriculture, but losses of land to
developed use occurs.10 Developed use is projected
to grow substantially over time, attracting land
from both forestry and agriculture and thereby
reducing, to some extent, the capacity of the forest
and agriculture sectors to mitigate GHGs through
actions on the land base.
Figure 3 4: Baseline Land Use Projections, FASOMGHG: 2010 2050 (Million acres)
2010
2050
• Cropland
• Pasture/Range
DTimberland (private)
D Developed Use
10The FASOMGHG projections for timberland out to 2050 are lower than those projected by the USDA Forest Service in their most
recent RPA projection (USDA Forest Service 2003, Ch. 2, Table 5) primarily because of differences in coverage—the latter
includes all 50 states, while the former includes the 48 contiguous states only. However, FASOMGHG projects a 9 percent loss of
timberland between 2010 and 2050, while the USDA Forest Service projects a 4 percent loss of timberland. The economic forces
captured by FASOMGHG suggest a more fluid change in land use than the USDA Forest Service methods.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-15
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
As indicated above, FASOMGHG projections for
declining cropland are consistent with recent trends
observed in the United States. Table 3-7 reports
data from the NRI, which tracks land-use change
across major categories from 1982 to 1997. The
biggest single change was in the area of cropland—
a net loss of about 44 million acres (10.4 percent of
the 1982 total). NRI data (not shown in the table)
indicate that three-quarters of the 1982 to 1997
cropland loss total was diverted to CRP lands
(about 33 million acres); the remaining lost crop-
land is net transfers to pasture and range, forest-
land, developed, and other uses. The CRP was
established to remove cropland from production
that is highly susceptible to erosion or otherwise
unproductive. In the scenarios throughout this
report, CRP land is assumed to remain permanently
at the initial level of 33 million acres.
Factors Underlying Land-Use Change Trends
For private lands in a market economy, land-use
decisions generally reflect each landowner's desire
to maximize the utility obtained from his or her
land by trying to maximize land profits (also called
land "rents"). These landowners may be very
responsive to changes in commodity output prices
and input prices and make land management
decisions to change the products they produce and
the inputs they use as prices vary. Other landown-
ers may place more emphasis on the nonmarket
services provided by their land such as rural
lifestyles, or wildlife habitat, more than maximiz-
ing the land's net income (Birch and Moulton
1997). These landowners may be less responsive
to constantly changing market signals than more
profit-oriented landowners. Over time these
market signals—including GHG market price
signals addressed in this report—may affect the
landowner's land-use decisions under changing
market and nonmarket conditions. Farmers may
adopt conservation tillage practices, establish
buffers along riparian corridors, and retire unpro-
ductive lands independent of, or in response to,
market incentives for GHG mitigation.
Price trends in forestry and agricultural commodi-
ties or technological advances in equipment and
land management options may be the largest
factors influencing land-use change for rent-driven
landowners. Figure 3-5 plots estimates of total
Table 3-7: U.S. Land-Use Change for Major Categories: 1982-1997
Million Acres
Land Cover/Use
1982
1997
Change
Percent
Cropland
Conservation Reserve Program (CRP)
Pasture
Rangeland
Forestland3
Other rural land
Developed land
Water areas and federal land
Total
420.6
0.0
131.9
416.4
403.0
49.6
73.2
447.9
1,942.6
376.7
32.7
119.9
405.7
406.6
51.1
98.2
451.8
1,942.6
-43.9
32.7
-12.0
-10.8
3.6
1.5
25.0
3.9
0.0
-1 0.4%
—
-9.1%
-2.6%
0.9%
3.0%
34.1%
0.9%
—
Source: USDA NRCS (2000).
a Forestland tracked by USDA, NRCS encompasses all productive timberland, as defined by USDA Forest Service, and reported in
Table 3-6, plus forestland that is not considered productive enough to be timberland.
3-16
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
factor productivity in U.S. agriculture over the last
half of the twentieth century,11 averaging 1.8
percent per year. However, from 1979 to 1999, the
average annual increase in productivity was about
2.3 percent.
During this period, real agricultural prices (i.e.,
net of inflation) have trended downward; net farm
income has stayed about even; and, as discussed
above, land devoted to agriculture has dropped.
Increases in agricultural productivity have reduced
the amount of land needed for agriculture, leading
to land retirement (CRP) and movement to pasture/
range, timberland, or developed uses. As shown in
Figure 3-6, the rise in forest-sector prices relative
to agricultural prices provides incentive for that
movement of land, along with increases in popula-
tion and real income.
Figure 3 5: Total Factor Productivity in U.S. Agriculture: 1949 1998
s
q
CO
x
V
T3
1.0
0.8
—/V
1948 1952 1956 1960 1964 1968 1972 1976 1980 1984 1988 1992 1996
Year
Source: Ball, Butault, and Nehring (2001).
Data for figure downloaded from http://www.ers.usda.gov/data/agproductivity/.
Figure 3 6: Forest and Agriculture Products Price Series
200
1982=100
I
1 Forest
'Agriculture
40
1926
1946
1966
Time (years)
1986
Source: U.S. Department of Labor, Bureau of Labor Statistics (Annual Series).
11 Total factor productivity measures the relative change in the ratio of total output produced to all inputs used. It is a comprehen-
sive measure of productivity and is a standard measure of technical efficiency in production.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-17
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Another significant driver of land-use change
is population growth. Population grew about 24
percent in the United States between 1980 and
2000 (Hobbs and Stoops 2002). Table 3-7 provides
evidence of population's effect on land use: devel-
oped land uses experienced the highest increase
between 1982 and 1997, with 25 million acres of
land undergoing development during that time
period, an increase of more than one-third.
Baseline GHG Projections
Table 3-8 presents the FASOMGHG baseline
projection of net GHG emissions from the U.S.
forest and agriculture sectors for decades 2010 to
2050 by specific activity group. The table reveals
that the sectors host a unique mix of activities.
Some activities, on balance, remove more GHGs
from the atmosphere than they emit (e.g., forest
carbon and, in some cases, agricultural soil carbon
sequestration). Some are pure emission sources
(e.g., CO2 emissions from fossil fuel use, agricul-
tural CH4 and N2O emissions). A small amount
of baseline biofuel (biomass) offsets is expected
to be generated in the form of ethanol substitution
for liquid fuels. The net atmospheric GHG effect is
negative (GHG removal), because these renewable
biofuels replace the burning of fossil fuels.
To summarize, the most important baseline
sectoral GHG effects over time are the following:
• The private forest sector is a net carbon sink,
absorbing more CO2 than it releases through
harvests and land-use change. The sink effect,
though, is projected to diminish in magnitude
over time, from 436 Tg CO2 per year in 2010 to
170 Tg CO2 per year in 2050. In the baseline,
there is some afforestation taking place in the
Table 3-8: Baseline Forest and Agriculture GHG Net Annual Emissions by Activity and Decade for the
United States: FASOMGHG Model: 2010-2050
Forest-sector (private) sources/sinks"
Afforestation
Forest management
Agriculture-sector sources/sinks (direct)b
Agricultural soil carbon sequestration
Agricultural CH4 and N2O
Sources/sinks from agriculture-energy
sector linkages0
Fossil fuel from crop production
Biofuel offsets
Combined forest- and agriculture-sector
net GHG emissions"
^vs 1 U
(436)
(114)
(322)
521
32
489
186
197
(11)
270
2020
(222)
92
(314)
513
10
503
189
200
(11)
479
2030
(145)
18
(163)
477
(83)
560
202
213
(11)
535
2040
(225)
4
(229)
449
(148)
597
218
229
(11)
442
2050
(170)
26
(196)
459
(167)
626
231
242
(11)
520
a Sum of afforestation and forest management.
b Sum of agricultural soil carbon sequestration and agriculture CH4 and N2O.
c Sum of fossil fuel from crop production and biofuel offsets.
d Sum of three categories above.
Notes: All quantities are in Tg CO2 Eq. per year. Negative (parenthesized) values are removals from the atmosphere (sinks). Positive
(nonparenthesized) values are emissions to the atmosphere (sources); decade means annual average value for that decade. Some
rounding error may occur.
3-18
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
first decade but not beyond that. Consequently,
future decades show losses in carbon accumu-
lated since the base year because of harvesting
of the afforested lands.12
• Net "direct" agricultural GHG emissions—the
sum of agricultural non-CO2 emissions and soil
carbon sequestration—exceed 500 Tg CO2 per
year in the baseline's first decade but eventually
decline. Non-CO2 emissions are projected to
rise steadily throughout the projection period,
but this rise in emissions is expected to be offset
by soil carbon sequestration, which starts as
a source but becomes a sink in later years. By
2050, agricultural soil carbon sequestration
draws even with forest carbon sequestration
at about 170 Tg CO2 per year.
• Net emissions from agriculture attributable to
energy production include CO2 emissions from
fossil fuel use in agricultural inputs offset by
biofuel production in agriculture. Together,
these factors are projected to account for 186 Tg
net CO2 per year in the 2010 decade, rising to
about 230 Tg CO2 per year in the 2050 decade,
a gain of about 25 percent.
• Combining all direct and indirect sources and
sinks in the combined forest and agriculture
sectors, the model baseline is somewhat vari-
able over time. The substantial drop in baseline
forest carbon sequestration over the first 2
decades causes a substantial increase in the
combined forest- and agriculture-sector net
GHG baseline emissions, essentially doubling
between 2010 and 2030 (270 to 535 Tg CO2 per
year). This GHG build-up reverses direction
after 2030, as carbon sequestration from both
forests and agricultural soils overtakes the rise
in sector GHG emissions.
Comparison of FASOMGHG Baseline GHG
Projection to Other Published Estimates
Several estimates exist of historic and projected
GHG trends in U.S. forestry and agriculture,
including those reported by EPA, USDA Forest
Service, and others. We review these estimates
here and compare them to the baseline used in
the FASOMGHG model.
Forest Carbon Sequestration
For forest carbon, we rely on two principal base-
line studies that have estimated past, current, and
projected carbon sequestration rates of American
forests:
• U.S. EPA, Inventory of U.S. Greenhouse Gas
Emissions and Sinks, 1990 - 2003 (EPA 2005)
• USDA Forest Service, Carbon Sequestration in
Wood and Paper Products (Skog and Nicholson
2000)
EPA GHG Inventory Baseline. The national
GHG inventory (EPA 2005) reports GHG emissions
and sinks in the United States since 1990. Table 3-9
shows the net flux in CO2 equivalents resulting
from forestry activities, including the amount of
carbon stored in harvested wood products. This
combined forest + wood products measure is the
most directly comparable to the FASOMGHG
forest carbon measure. Together, the forest carbon
sink components account for over 90 percent of all
terrestrial carbon sequestration in the inventory;
the remaining portion comes from agricultural
soil carbon. Carbon contained in wood products
constitutes about one-quarter to one-third of the
total forest carbon sequestration total.
The total forest carbon flux reported in the EPA
inventory declined steadily from 1990 to 2000. In
1990, the sector generated a net sink of nearly 950
Tg CO2 Eq. per year, but this declined by about 200
Tg per year by 2000. Two-thirds (137 Tg CO2 Eq.
per year) of the decline in sequestration from 1990
to 2000 is attributable to a change in the methods
used to estimate SOC between the two periods.
The remaining third (64 Tg) is attributable to a
reduced rate of afforestation, which was quite high
in the late 1980s and early 1990s partly because of
public conservation programs such as the CRP.
12The base year for these simulations is 2000. Model results are reported for the period 2010 to 2050 (see Chapter 4). Some of the
carbon losses from "afforestation" are based on lands afforested in the 2000 decade.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-19
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Table 3-9: Net Annual CO2 Flux from U.S. Forest
Carbon Stocks: 1990 and 2000, EPA
Inventory Quantities (in Tg CO2 per
year)3
Component
1990
2000
Forest (739) (537)
Above ground (396) (400)
Below ground (77) (78)
Dead wood (74) (45)
Litter (67) (26)
Soil organic carbon (SOC)b (125) 12
Harvested wood (210) (211)
Wood products (48) (59)
Landfilled wood (162) (152)
Total net annual flux (949) (748)
Difference in net flux: 2000 vs. 1990 201
Difference, net of SOC 64
Source: EPA (2005).
a Negative (parenthesized) values are removals from the
atmosphere (sinks). Positive (nonparenthesized) values are
emissions to the atmosphere (sources).
b SOC differences are primarily due to changes in estimation
methods.
USDA Forest Service Forest-Sector Baseline.
The estimates in the EPA inventory report recent
historical trends since 1990, but future projections
are necessary for comparison against the
FASOMGHG baseline. EPA estimates for the forest
sector were derived collaboratively with the USDA
Forest Service, using USDA Forest Service models
referenced above (e.g., FORCARB). Therefore,
we turn to a recent study by USDA Forest Service
researchers that estimates national levels of forest
carbon sequestration into the future to provide a
consistent framework for comparison.
In 2000, the USDA Forest Service produced a
comprehensive assessment of national forest
carbon stocks and flows. Within that report, a
chapter by Skog and Nicholson (2000) presents
a set of projections for the period 1990 to 2040 that
can be matched to the forest carbon categories
reported by EPA above. The USDA Forest Service
projections are presented in Table 3-10. According
to those estimates, U.S. forest carbon sequestration
exceeded 1.2 Gt CO2 per year in 1990, at which
point a steady decline is projected to extend but
taper off through the middle of the 21st century.
The forest sink is projected to decline about 360
Tg CO2 per year (30 percent) from 1990 to 2040.
But virtually all of that decline is found in the 1990
to 2000 decade, mirroring the drop reported in the
EPA GHG inventory for that same time period. The
projected annual decline in forest carbon seques-
tration from 2000 to 2040 is just 5 percent.
Table 3-10: Projected Net CO2 Flux from U.S. Forest Carbon Stocks: 1990-2040, USDA Forest Service
Estimate
Net CO2 Flux (Tg CO2 per year)
1990
2000
2010
2020
2030
2040
Change in forest carbon stocks
Changes in harvested wood carbon stocks
Change in products in use
Change in landfills
Total change in stock of carbon
1,006
218
96
123
1,224
694
211
92
119
905
705
235
90
145
939
646
250
94
156
896
609
261
89
172
870
591
270
84
186
861
Source: Table 5.7 in Skog and Nicholson (2000).
3-20
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
Comparison of Baseline Projections: USDA
Forest Service and FASOMGHG. We now
compare FASOMGHG's forest carbon baseline
projections with projections for the corresponding
time period by USDA Forest Service (Skog and
Nicholson 2000). The comparison is illustrated
in Figure 3-7.
Before proceeding with the comparison, we note
several important points. First, the projection time
periods do not exactly match: the USDA Forest
Service projections run from 1990 to 2040, and
FASOMGHG's projections run from 2010 to 2050.
Therefore, the most meaningful comparisons are
from 2010 to 2040. Second, in Tables 3-9 and 3-10
note the difference in the quantities between the
EPA and USDA Forest Service estimates for 1990
and 2000. The 2000 value reported in the USDA
Forest Service report is more than 150 Tg CO2
higher than the EPA inventory estimate. Much
of this difference is due to the methods-based
adjustment in soil carbon estimates between
1990 and 2000 that is reflected in the EPA (2005)
estimate but not in the Skog and Nicholson (2000)
estimate. Because this soil adjustment is method-
ological in nature, we recalibrated the Skog and
Nicholson projections to be more consistent with
the EPA projection using the revised methodology.
We did that by adjusting the USDA Forest Service
projection downward to match the EPA estimate
for 2000 (748 Tg CO2) and then allowing the USDA
Forest Service projection for 2000 to 2040 to
pertain beyond that.
Third, the USDA Forest Service projections are
for all forestland in the United States (private and
public), while the FASOMGHG projections are for
private land only. Although the inventory data for
public forestland are somewhat incomplete, these
forests are estimated to provide a substantial net
carbon sink in the United States (Heath 2000).
That essentially explains the large gap between
the FASOMGHG and USDA Forest Service lines
in Figure 3-7.
Putting aside the public lands gap in Figure 3-7,
both sets of projections show a similar pattern,
namely that the forest carbon sink is projected
to decline over time. The decline is a bit more
pronounced in FASOMGHG, reflecting differences
in the methods used to create the projections.
FASOMGHG uses economic principles and
dynamic optimization methods to allocate
resources across time, while the system used
by Skog and Nicholson is not as explicitly driven
by economic models of intertemporal economic
behavior. However, both sets of projections are
consistent in their assessment that under BAU
Figure 3 7: Comparison of Projected Baseline Carbon Sequestration Trends in U.S. Forests:
FASOMGHG vs. USDA Forest Service Model
1,
8
o
o
D)
1,000
750
500
250
USDA Forest Service
(2000), calibrated to
EPA (2005), public
and private lands
FASOMGHG (private
lands)
1990
2000
2010
2020
Year
2030
2040
2050
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-21
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
conditions, the rate of CO2 sequestration in U.S.
forest ecosystems is slated to decline over time.
Therefore, absent any policy interventions or
unforeseen changes in natural, economic, or
institutional phenomena, the forest sector's role
in partly offsetting the country's GHG emissions
will diminish.
To summarize, forests make up the lion's share
of current terrestrial sequestration in the United
States and are a net sink because the amount of
CO2 currently taken up through photosynthesis
and stored in biomass, soils, and products exceeds
the amount released through harvesting and
natural disturbances. This is the result of recent
land-use trends, which show a net movement of
land from agriculture to forests, and an age class
structure of U.S. forests favoring younger, faster-
growing trees. However, under BAU, these land-
use conversions are not expected to occur at the
same rate. Additionally, timberland is projected to
be diverted to developed uses over the projection
period, thereby leading to forest carbon losses.
Taking these factors together, future sequestration
rates in the U.S. forest sector are expected to
decline below the rates we are now experiencing
in the absence of additional forest carbon seques-
tration activities.
Agricultural Soil Carbon Sequestration
As was shown in Table 3-8, FASOMGHG projects
agricultural soil as a net emitter of CO2 in the early
periods (about 30 Tg CO2 in 2010) and as a signifi-
cant sink in later years (nearly -170 Tg CO2 in 2050),
thereby tipping the sector's carbon balance toward
sequestration by about 200 Tg CO2 during this
time period.
Although there are no published projections of
future baseline agricultural soil carbon sequestra-
tion to compare with the FASOMGHG projections
for 2010 to 2050, one can compare the 2010 projec-
tion—a small source of +32 Tg CO2/year—with the
most recent estimate (for data year 2003) reported
in the U.S. GHG inventory (EPA 2005)—a small
sink of -7 Tg CO2/year. This gap reflects a differ-
ence between methods used in FASOMGHG (i.e.,
CENTURY model) and methods used in the EPA
inventory (IPCC default factors with U.S. data),
and assumptions on short-run baseline adoption
of practices to sequester agricultural soil carbon.
The FASOMGHG model reveals a pattern of low
adoption of sequestration practices (predominately
reduced tillage) in the early years of the projection
but robust adoption in later years in response to
projected changes in the underlying market and
technological conditions. The EPA inventory
estimates may reflect some adoption occurring
sooner than projected in the FASOMGHG model.
Other differences in underlying phenomena
involving soil sequestration also may be occurring,
such as the rate of cropland conversion to grass-
land and changes in nontillage soil management,
including the addition of manure amendments.
Non-CO2 GHG Emissions in Agriculture
According to the national GHG inventory report
(EPA 2005), agricultural practices directly account
for about 6 percent of all GHG emissions in the
United States, primarily in the form of CH4 and N2O.
These non-CO2 GHG emissions from agriculture
totaled about 433 Tg CO2 Eq. in 2003 (see Table 3-11).
As discussed earlier in this report, the primary
sources of these GHGs in agriculture are fertilizer
applications on croplands, enteric fermentation,
manure management, and rice cultivation. Residue
burning is also a small source of non-CO2 gas
emissions from agriculture. According to the
national GHG inventory report, agriculture account-
ed for about 30 percent of all CH4 emissions and 72
percent of all N2O emissions in the United States.
Table 3-11 presents recent levels of agriculture
non-CO2 GHG emissions. The trends presented
in Table 3-11 show a fairly slight (1.6 percent)
increase in sector emissions between 1990 and
2003. Although they have increased, agricultural
emissions have done so at a slower rate than total
U.S. GHG emissions (EPA 2005).
Although the EPA inventory estimates are historic,
a recent paper by Scheehle and Kruger (in press)
provides projections for non-CO2 GHG emissions
out to 2020. Those projections are compared to the
FASOMGHG projections in Figure 3-8 and are
found to match rather well. The magnitudes of the
3-22
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
estimates are within 5 percent of each other and
both show a rising trend in non-CO2 emissions
over the next several decades.
Sources/Sinks from Agriculture-Energy
Linkages
As reported in Table 3-8, a sizeable portion of
the sector's total emissions originate from CO2
released in fossil fuel combustion embodied in
the energy to produce agricultural inputs. As
described above, this not only includes on-farm
use of fuels in farm machinery, but also the
upstream energy use in the production of inputs,
such as the amount of energy used to produce
fertilizer. This is a more expansive definition
of agricultural CO2 emissions than others have
employed and therefore there are no direct
Table 3-11: Non-CO2 GHG Emissions from Agriculture (Tg
Gas/Source
CH4
Enteric fermentation
Manure management
Rice cultivation
Agricultural residue burning
N2O
Agricultural soil management
Manure management
Agricultural residue burning
1990
156.9
117.9
31.2
7.1
0.7
269.6
253.0
16.3
0.4
1997
163.0
118.3
36.4
7.5
0.8
269.8
252.0
17.3
0.4
1998
164.2
116.7
38.8
7.9
0.8
285.6
267.7
17.4
0.5
C02 Eq.):
1999
164.6
116.8
38.8
8.3
0.8
261.3
243.4
17.4
0.4
EPA GHG
2000
162.0
115.6
38.1
7.5
0.8
282.1
263.9
17.8
0.5
Inventory, 1990-2003
2001
161.9
114.5
38.9
7.6
0.8
275.6
257.1
18.0
0.5
2002
161.5
114.6
39.3
6.8
0.7
270.9
252.6
17.9
0.4
2003
161.8
115.0
39.1
6.9
0.8
271.5
253.5
17.5
0.4
Non-CO, GHG Emissions Total 426.5
432.8
449.8
425.9
444.1
437.5 432.4
433.3
Note: Totals may not sum due to independent rounding.
Source: These numbers are taken from EPA (2005).
Figure 3 8: Comparison of Projected Baseline Non CO2 GHG: FASOMGHG vs. Scheehle and Kruger
(in press)
300 -
2000
2010
Year
1 Scheehle and Kruger (in press)
2020
2030
• FASOMGHG
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
3-23
-------
CHAPTERS • MODELING FRAMEWORK AND BASELINE
comparisons that can be made to the FASOMGHG
estimate. The closest comparison one can make is
to the 2005 EPA GHG inventory, which shows CO2
emissions from agricultural equipment of about
41 Tg CO2 per year in 2003 (EPA 2005, Table 3-36
in Annex 3-2).
Applying FASOMGHG for the Purposes
FASOMGHG evaluates the joint economic and
biophysical effects of GHG mitigation policies in
the U.S. forest and agriculture sectors. The model
considers most major GHG mitigation options and
GHG flows in the two sectors over an extended
time period. As an economic model, FASOMGHG
ensures consideration of the effects of policy
initiatives on resource flows and economic activi-
ties within and across the forest and agriculture
sectors over time. It has sufficient detail to answer
questions about which activities are economic, how
much GHGs are reduced by their adoption, and
where and when the actions are likely to occur.
Interpretation of the model results can provide
insights into how and why these activities and
GHG effects occur.
FASOMGHG and its component models have been
extensively peer reviewed.14 The model is consis-
tent with modern economic theory, agronomy, and
ecology. FASOMGHG is empirically grounded
with base period data (ca. 1990 to 2000) tied to
published projections of key data and parameters
for simulation of future scenarios.
The comprehensiveness, detail, theoretical consis-
tency, and empirical grounding of FASOMGHG
make it suited for policy analyses of GHG policies,
including the introduction of GHG (sometimes
called carbon or CO2) prices, GHG quantity goals,
and nuanced combinations thereof. Like any
model, some abstraction of real-world complex
details is necessary to make the problem tractable,
which can hinder the flow of some information.
Therefore, one may want to focus more on the
broad and subtle patterns found in the model
results and what they mean for GHG policy, rather
than on specific estimates of a GHG or economic
effect at a certain point in place and time.
14 For a selected listing of publications using FASOMGHG and its predecessor models (ASM andFASOM),
see http://agecon2.tamu.edu/people/faculty/mccarl-bruce/papers.htm.
3-24
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4
Mitigation Potential:
Comprehensive Scenarios with
All Activities and All GHGs
Chapter 4 Summary
Mitigation results are presented for all forest and agricultural activities and all GHGs under
constant and rising GHG price scenarios over a range of $1 to $50 per t CO2 Eq. (or roughly $4 to
$184 per t C Eq.). Mitigation quantities are reported as changes from FASOMGHG's baseline. Low
GHG price incentives have little effect on land-use change, but higher prices can induce substantial
land-use change from agriculture to forestry and changes in practices within sectors. The price level
affects the optimal portfolio of mitigation strategies. Carbon sequestration from agricultural soil
practices and forest management dominates at lower GHG prices and in the near term. These two
options produce about 90 percent of all mitigation in the earlier years, but these annual sequestra-
tion effects diminish by 2055. Afforestation dominates mitigation at higher prices in the early to
middle years. However, carbon sequestered in afforestation is reversed by 2055, at which time the
planted forests become a net CO2 source. At the highest prices and in the later years, biofuels are a
dominant strategy.
Timing effects vary depending on the GHG price scenario. In the constant-price scenarios, GHG
mitigation declines over time, as landowners react early to incentives. Declining rates of mitigation
are the result of carbon saturation (reaching a new equilibrium), harvests, and the conversion of
forests back to agriculture. Despite these declining annual mitigation rates, cumulative mitigation
steadily increases. In the rising-price scenarios, GHG mitigation increases overtime as landowners
are assumed to fully recognize that prices will rise and therefore employ some mitigation actions
later. Mitigation potential has a regional distribution. The South-Central, Corn Belt, and Southeast
regions possess the largest GHG mitigation potential, while the Rockies, Southwest, and Pacific
Coast regions generate the least.
Chapter 3 describes the modeling frame-
work of FASOMGHG and its projected
baseline of GHG emissions and sinks in
U.S. forestry and agriculture. This chapter pres-
ents FASOMGHG mitigation results as changes
from the baseline, in terms of additional carbon
sequestration and GHG reductions. Mitigation
results are presented for a range of hypothetical
scenarios that include both constant and rising
economic incentives for GHG mitigation over time.
More specifically, results from the GHG mitigation
scenarios show management and land-use changes,
average annual GHG mitigation for selected years
(focusing on the next few decades), cumulative
GHG mitigation over time, results by region, results
by individual mitigation option, and a brief over-
view of key environmental co-effects. The emphasis
here is on identifying and quantifying GHG miti-
gation opportunities at various economic values of
GHGs, not on simulating a specific policy.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-1
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Mitigation Responses under Various
GHG Mitigation Scenarios
This section estimates net GHG emissions from
U.S. forestry and agriculture, reported as changes
from the baseline levels, through a combination
of sequestration and emission reduction strategies.
The primary approach evaluated throughout
this report is the assignment of a price for GHG
emissions and sequestration. Under such pricing,
landowners or other economic agents would
receive payments for increasing sequestration and
reducing emissions and would make payments for
increasing emissions or reducing sequestration.
The actual mechanism of providing GHG incen-
tives and disincentives for participants specifically
is not addressed here. The basic principle in the
GHG price analyses below is that GHG prices
provide incentives for increasing sequestration
through land-use change, forest management,
conservation tillage, and other forms of land
management, and for decreasing emissions
through land-use change (e.g., deforestation),
harvesting, input use, and processes that generate
non-CO2 GHGs.
Varying the prices of GHGs in the FASOMGHG
model of the forest and agriculture sectors allows
for an evaluation of the total GHG mitigation
potential from these sectors at different economic
incentive (price) levels and identifies the activities
and regions that comprise the most cost-effective
portfolio of mitigation options. Proposing or
designing specific climate mitigation policies
for these sectors is beyond the scope of this report.
Thus, the section continues with a description
of hypothetical core price scenarios for GHG
emissions and sequestration. This approach is
consistent with numerous modeling efforts con-
ducted in the recent past that have examined GHG
mitigation responses across countries, time, and
sectors to hypothetical GHG price scenarios.1
Following the scenarios description, the section
presents mitigation results from the FASOMGHG
model. Variations on these core price scenarios are
presented in subsequent chapters.
Boxes 4-1 and 4-2 detail reporting conventions
used throughout the next few chapters with
respect to measurement units and mitigation
quantities across time periods.
Scenarios Description: Constant and Rising
Incentives for GHG Mitigation
The mitigation analysis begins by stipulating a core
set of scenarios that simulate the effects of setting
a value for GHGs and modeling the subsequent
effect on economic behavior and GHG emissions
and sequestration.
Constant-Price Scenarios
The core price scenarios are described in Table 4-1
and are divided into two groups. The first group
includes the constant-price scenarios, which
evaluate GHG price levels ranging from $1 to
$50 per tonne of CO2 equivalent (t CO2 Eq.) but
assumes that the prices remain constant in real
(inflation-adjusted) terms over time. Because many
climate-modeling analyses use carbon (C), rather
than CO2, as the unit of measure, Table 4-2 presents
the carbon price equivalent to the CO2 prices. The
purpose of evaluating a range of GHG prices is
to see not only how the total level of mitigation
changes over the price range, but how the composi-
tion by activity and region changes as well.
Box 4-1: Measurement Units Reported in the
Analysis
1 The units of exchange for all GHGs are tonnes (t) of
CO2 equivalent (Eq.): 1 tonne (metric ton) = 1,000 kg
= 1 Megagram (Mg) = 1.102 short tons = 2,205 Ibs.
' CH4 and N2O are converted to CO2 Eq. with GWPs
from the IPCC (1996) Second Assessment Report
(see Box 1-1 in Chapter 1).
1 Most mitigation results in this and subsequent
chapters are given in teragrams (Tg) of CO2 Eq. 1
Teragram = 1 million tonnes.
1 For a sample of modeling efforts evaluating the effects of broad GHG incentive analyses, consult Web sites for the Stanford
Energy Modeling Forum (EMF) (http://www.stanford. edu/group/EMF/publications/index.htm), the MIT Joint Program on the
Science and Policy of Climate Change (http://web.mit.edu/globalchange/www/reports.html), and The Pew Center for Global
Climate Change (http://www. pewclimate.org/policy_center/reports/) among others.
4-2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Box 4-2: Methods Used for Reporting GHG Mitigation Results at Different Points in Time
Annual averages: Present the average level of GHG
reductions represented in FASOMGHG for a given year.
For the purposes of this report, the annual values for 3
specific years—2015, 2025, and 2055—are used to
represent results in the short, intermediate, and long
runs. These years represent the midpoint of the
decades 2010, 2020, and 2050 tracked in the model
and are annual averages for the decades.
Cumulative: Reports results as the cumulative GHG
mitigated over the full projection period or period
specified. This value is the amount of GHG mitigated
in year n plus the total amount mitigated in year (n - 1)
+ (n - 2) + (n - 3)... back to the beginning year of the
simulation (2010). Although specific options may
increase emissions compared to the baseline, the
cumulative effect may still be a net GHG reduction
as a result of the reductions from the full suite of
mitigation options.
Annualized quantities: Because mitigation effects
can vary tremendously over time, a concise summary
metric is needed to convey the GHG mitigation potential
over a given time period. The metric used for these
purposes in this report is the annualized equivalent
value GHG mitigation quantity. The annualized equiva-
lent refers to the equivalency between the net present
value of all GHG mitigation over a given projection
period (typically the full horizon, 2010 to 2110, but shorter
time horizons can be considered)—accounting for
variable GHG gains and losses over time—and receiving
a fixed quantity of GHG mitigation each year for the
same projection period. By using net present value
concepts, the annual GHG effects are time discounted;
therefore, near-term effects are weighted more heavily
than those in later time periods. (The rationale for such
an approach is discussed in Herzog et al. [2003].) The
discount rate used is 4 percent per year. More informa-
tion on this metric is provided in Box 4-5.
Table 4-1: Core Price Scenarios
Initial Price in 2010
($/t C02 Eq.)
Annual Price Growth
Price Cap
Constant Prices
Rising Prices
$1
$5
$15
$30
$50
•
$3
$3
$20
0
0
0
0
0
•
1.5%/yr
4%/yr
$1.30/yr
None
None
None
None
None
•
None
$30
$75
Table 4-2: CO2 and C Price Equivalents
($
CO2 Price
per t C02 Eq.)
$1
$3
$5
$15
$20
$30
$50
$75
C Price
($ per t C Eq.)
$3.67
$11.01
$18.35
$55.05
$73.40
$110.10
$183.50
$275.25
Note: One unit of C equates to 3.67 units of CO2.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-3
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Rising-Price Scenarios
The second group of scenarios in Table 4-1
addresses rising GHG prices, wherein an initial
price is asserted beginning in Year 2010, as well
as a rate of increase over time. These scenarios
provide a means to examine whether the incentive
for delayed action to capture mitigation at higher
future prices is quantitatively important in these
sectors. Figure 4-1 shows the price trajectories
associated with each of the three rising-price
scenarios, illuminating the differences in the rate
of increase and price levels attained.
The first two rising-price scenarios have a modest
initial price of $3/t CO2 Eq., rising alternatively at
1.5 and 4 percent per annum over the time period.
The price caps out at $30It CO2 Eq. under the 4
percent price rise scenario. The third scenario
commences at a price of $20It CO2 Eq., rising at
$1.30 per year, capping out at a price of $75. This
third price scenario roughly matches a fairly
aggressive price path considered by modeling
efforts tied to the Stanford University EMF (http://
www.stanford.edu/group/ EMF/home/index.htm).
Price caps are introduced to keep carbon prices
from reaching seemingly unrealistic levels and are
in accordance with other scenarios tested in past
research. For further discussion of rising price
scenarios, see van't Veld and Plantinga (2005).
The model is initially run to reflect comprehensive
coverage. Comprehensive means that all forestry
and agricultural activities and all GHGs (CO2,
CH4, N2O) represented in FASOMGHG are subject
to the GHG payment scenarios. These results,
in essence, help identify the competitive potential
of individual mitigation options and of the aggre-
gate U.S. forest and agriculture sectors for GHG
mitigation. See Box 4-3 for a description of techni-
cal, economic, and competitive potential as they
relate to assessing GHG mitigation. Later, the
report considers a more refined set of scenarios
that are less comprehensive and more selective
in coverage.
The FASOMGHG model is run in decadal time
steps for the time period 2010 to 2110. Because
there is greater uncertainty in model projections
beyond the first several decades, the analysis
results focus primarily on selected years: 2015,
2025, and 2055. Longer-term results are presented
to highlight the unique temporal dynamics of
carbon sequestration mitigation strategies in
the forest and agriculture sectors. The following
discussion focuses first on mitigation results for
the constant-price scenarios and then turns to
results for the rising-price scenarios.
Figure 4 1: Price Trajectories for Rising Price Scenarios
2000
2020
2040
$3 @ 1.5% per yr
2060
Year
1 $3 @ 4% per yr
2080
2100
2120
1 $20® $1.30 peryr
4-4
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Mitigation Response to Constant GHG
Price Scenarios
The mitigation responses to the constant GHG price
scenarios are presented in the following order:
• land-use and land management effects,
• total national GHG mitigation quantities for
selected years,
• total cumulative GHG mitigation over time,
• GHG mitigation by individual forestry and
agricultural activities,
• GHG mitigation by region, and
• non-GHG environmental co-effects.
Box 4-3: Technical, Economic, and Competitive
Potential of a GHG Mitigation Option
Example: U.S. agricultural soil carbon
sequestration potential
0)
o
CL
CD
I
CD
Competitive Economic^
Potential Potential
Technical
Potential
Soil Carbon Sequestration
Source: McCarl and Schneider (2001).
The technical potential reflects the maximum biophys-
ical potential for GHG mitigation if all resources were
committed to this objective without regard to cost.
The economic potential incorporates the cost of
mitigation options by showing that increasing levels of
compensation are necessary to procure higher levels
of GHG mitigation from the activity. The economic
potential can fall well within the technical potential at
price ranges considered in this analysis. Finally, the
competitive potential reflects the interaction of the
GHG mitigation activity with all other activity in the
forest and agriculture sectors.
For example, while the economic potential shows that
agricultural soil carbon sequestration becomes more
profitable at higher prices, the competitive potential
recognizes that other mitigation options within the
sectors (such as afforestation and biofuels) also
become more profitable at higher prices. Therefore,
some of the economic potential for agricultural soil
carbon sequestration is diverted to other more
profitable options within forestry and agriculture at
higher GHG prices.
A summary of the results that unfold under the
constant-price scenarios is presented in Box 4-4.
Land-Use and Land Management Effects
The GHG price incentives alter the economic
returns to land and can thereby affect the way that
land is allocated across uses. Figure 4-2 illustrates
this by showing differences in land use in Year
2025 simulated by variations in the GHG price.
The largest impact is on private timberland, which
increases from 315 million acres (128 million ha) in
the baseline ($0 price) to about 427 million acres
(173 million ha) at the $50/t CO2 Eq. price, reflect-
ing the prominent role of afforestation in the
higher price scenarios. The gain in timberland
comes at the expense of losses in both cropland
and pastureland. However, this gain in timberland
may be temporary. As shown in Figure 4-3, the
large increase in timberland at the beginning of
the period brought about by a high GHG price
($50/t CO2 Eq.) dissipates over time as the total
Box 4-4: Summary of Constant GHG Price
Scenario Results
The mitigation responses to the constant GHG price
scenarios are summarized here and presented in
detail in the main text and in Table 4.A.1 in the
appendix:
• The lower GHG prices have little effect on land-use
change. Starting at the $15/t CO2 Eq. (or $55/t C
Eq.) price, however, appreciable effects on cropland
(decline) and timberland (increase) start to material-
ize. It is not until the highest prices that pastureland
begins to decline and biofuel lands increase.
• In the first decade, total national GHG mitigation is
low at the low GHG prices —121 Tg CO2 Eq./year
(or 33 Tg C Eq.) at the $1 CO2 price ($4/t C). This
would offset about 2 percent of total national
GHG emissions. However, under the highest price
scenario ($50), 1,500 Tg CO2, or over 21 percent of
the current national GHG emissions total, could be
mitigated.
• Forest management and soil carbon sequestration
are dominant at the lower GHG prices. At a $5 CO2
price, these activities account for 86 percent (260
Tg CO2 Eq., or 71 Tg C Eq.) of total mitigation by
2015.
• Afforestation is the dominant mitigation activity at
the higher GHG prices. At $50, 877 and 1,296 Tg
CO2 Eq. (or 239 and 353 Tg C Eq.) are mitigated by
2015 and 2025, respectively.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-5
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
area of timberland reverts back to baseline levels
after several decades. This reversion of lands to
baseline conditions is driven by the fact that, at
some point, the economic returns from converting
lands back to agriculture are higher compared to
keeping lands tied up in forestry. Moreover, there
continue to be exogenous demands for land to be
used for developed uses, which can divert land that
otherwise may be allocated to forests. Thus,
reversals occur in both land use and accumulated
carbon benefits.
In addition to altering the allocation of land uses,
GHG prices can also affect how land within a
major use is managed. Table 4-3 shows the area
of land converted from conventional crop tillage
to reduced tillage under the baseline and GHG
price scenarios over time.
In the baseline, FASOMGHG projects a fair
amount of new reduced tillage by 2015—20 million
acres (8 million ha)—and this amount grows over
time to more than 30 million acres (12 million ha)
Figure 4 2: Land Use in 2025 at Different GHG Price Levels
• Cropland
• Pasture
DTimberland (private)
DBiofuel Land
$1
*Baseline
Notes: $ represent price per tonne, CO2 Eq.
Quantities are in million acres.
$5 $15
GHG Price
$30
$50
Figure 4 3: Timberland Area over Time: $50/t CO, Eq. vs. Baseline
500
— 400
TO
2010
2090
• Baseline
• $50/Mg
4-6
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Table 4-3: Acreage Converted from Conventional Tillage to Reduced Tillage under Baseline and GHG
Prices: U.S. Total (Million acres)
Year
From Conventional Tillage to ...
GHG Price ($/t CO 2, constant over time)
Baseline
$1
$5
$15
$30
$50
2015
Conservation tillage
Zero tillage
Total reduced tillage
2025
Conservation tillage
Zero tillage
Total reduced tillage
2055
Conservation tillage
Zero tillage
Total reduced tillage
10.5
9.8
20.4
20.3
5.4
25.7
27.5
3.6
31.0
48.5
40.6
89.2
6.4
8.0
14.4
6.1
6.6
12.7
Million
31.4
111.7
143.1
0.3
4.9
5.3
0.1
3.1
3.2
Acres3
2.1
153.8
155.9
0.0
3.2
3.2
6.2
2.0
8.2
0.4
144.6
145.0
0.1
4.2
4.2
0.0
3.0
3.0
0.6
129.3
129.9
0.1
3.2
3.3
0.0
0.4
0.4
Baseline acres are the projection of tillage change under no GHG mitigation scenario. Acres for the GHG price scenarios are
absolute values, rather than differences from the baseline (note: many other estimates in the report are the latter).
by 2055. However, the amount of cropland converted
to reduced tillage rises dramatically under GHG
pricing, ranging from about 90 to 155 million acres
(36 to 63 million ha) by 2015. The latter number is
almost half of the nation's cropland base. Most of
this land goes into zero tillage ("no-till") practices.
This is especially pronounced at the higher GHG
prices, for which the extra financial gain from
reducing tillage further is most pronounced. Note
that the decline in tillage conversion after 2015
does not mean that reversion to conventional
tillage is occurring. Rather, it means that there are
fewer acres converting from conventional tillage to
conservation or zero tillage at that time, primarily
because most of these conversions have already
occurred in previous periods.
However, note that the total reduced tillage
acreage is highest at the $15 GHG price. Reduced
tillage acreage is lower under the $30 and $50
prices because the amount of total cropland is
projected to decline as land is diverted from crop
production to forests and biofuels at the two
higher prices, as shown in Figure 4-2. This relative
decline in tillage adoption at the highest prices
underscores the differences in economic and
competitive potential referenced in Box 4-3.
The introduction of GHG prices also induces
changes in forest management. Figure 4-4 illus-
trates the effects of different GHG prices on the
average rotation (harvest) age of existing timber
stands and the average management intensity of
timber stands that are reforested after harvest.
Chapter 2 discusses how GHG prices can extend
harvest rotation ages; Figure 4-4 gives empirical
evidence of this effect. Higher GHG prices tend
to lengthen the rotation age, although the effect is
not dramatic. The projected baseline (national)
average rotation age is about 56 years for the 2015
period. This rises to about 62 years at a price of
$50It CO2. Management intensity is indexed on a
scale of 1 to 4; 4 is the most intensive form of forest
management (e.g., site preparation, fertilization,
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-7
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
thinning, prescribed burns), and 1 represents
essentially no active management. Figure 4-4
shows that GHG prices raise management inten-
sity because the additional management generates
additional carbon.
Total National GHG Mitigation Quantities
for Selected Years
Figure 4-5 presents total national results for the
constant-price scenarios in terms of annual GHG
mitigation achieved for the focal years 2015, 2025,
and 2055. More detail on the contribution of
specific activities to the national mitigation total
for these key years can be found in Table 4.A.1 in
the appendix to this chapter.
As expected, the total amount of GHGs mitigated
by the forest and agriculture sectors rises with the
size of the economic incentive. In 2015, annual
mitigation totals for the forest and agriculture
sectors range from fairly modest at the $1 price
Figure 4 4: Effect of GHG Prices on Forest
Management Variables, 2015
Average Rotation Age
$50
$30
$15
$5
$1
Baseline
0 10 20 30 40 50 60 70
Average Management Intensity
$50
$30
$15
$5
$1
Baseline
0.0
1.0
2.0
3.0
(121 Tg CO2 Eq. per year) to substantial at the $50
price (about 1,500 Tg CO2 per year). These quanti-
ties are, respectively, just under 2 percent and just
under 22 percent of 2003 GHG emissions for the
United States (EPA 2005), the latter of which could
clearly be a substantial contribution to aggregate
national mitigation potential, although at that price
($50/t CO2 Eq. or $183.50/t C Eq.), mitigation
options from other sectors could be substantial
as well.
Note that the annual mitigation quantities rise
between 2015 and 2025, particularly at the higher
prices for which forest carbon sequestration from
afforestation—which takes some time to culmi-
nate—plays a more significant role in the mitiga-
tion portfolio, as discussed below. The mitigation
potential is generally lower in 2055 than in 2025 or
2015, reflecting the saturating and reversal effects
of sequestration options referenced above. More
discussion of the time element of mitigation
options in these sectors now follows.
Figure 4 5: National GHG Mitigation at
Representative Years by Price
(2015, 2025, and 2055)
Quantities are in Tg CO2 Eq. per
year net emissions reduction below
baseline.
2000
$50
$15
$10
$5
$1 GHG Price
($/t C02 Eq.)
4-8
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Total Cumulative GHG Mitigation
Over Time
Given the unique dynamics of carbon sequestra-
tion, it is especially important to look at cumulative
GHG mitigation results over time. In a given
year, a specific mitigation option can produce an
increase or reduction in GHG emissions relative
to the baseline. Reporting the results annually
may therefore hide the cumulative effect of the
mitigation options over time. The long-term
emission reductions and sequestration are more
important than short-term fluctuations when
addressing climate change issues.
Figure 4-6 shows cumulative GHG effects over
the entire projection period for the $15 and $30
per t CO2 Eq. constant price scenarios, respectively.
After several decades, some reversal of carbon
sequestration occurs as soil carbon equilibrium
points are reached and carbon reversals occur
through timber harvesting and reversion of
afforested lands back to agriculture. Afforestation
efforts early on in the period accumulate for
several decades as the newly planted trees seques-
ter carbon. Then, as the trees are harvested in the
future, CO2 is released again into the atmosphere,
reversing some of the cumulative carbon built up
Figure 4 6: Cumulative GHG Mitigation over Time
Quantities areTg CO Eq. cumulative net emissions reduction below baseline.
$15/t CO. Eq. Constant Real Price
D Biofuel offsets
D Crop management FF mitigation
D Ag CH4 and N2O mitigation
• Forest management
D Afforestation
• Agricultural soil C sequestration
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
$30/t CO. Constant Real Price
100,000
D Biofuel offsets
D Crop management FF mitigation
D Ag CH4 and N2O mitigation
• Forest management
D Afforestation
• Agricultural soil C sequestration
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
*Note differences in the quantity range on the vertical axis of each diagram.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-9
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
over time. Cumulative agricultural soil carbon
sequestration rises, then stabilizes after several
decades as the carbon benefits of reduced tillage
practices saturate. Forest management shows a
saturating and slight reversal effect as well.
These patterns highlight an important difference
between the duration of sequestration relative
to other mitigation options within the forest
and agriculture sectors. While the sequestration
options display saturation and impermanence, the
fossil fuel CO2 and non-CO2 emission reduction
options essentially do not. The latter reductions
are considered more permanent, because the
avoidance of an emission does not create the
same biophysical diminishing returns and risk
of re-release as sequestration.2 Differences
between the cumulative contribution of seques-
tration and nonsequestration options widen
over time and are particularly pronounced in
the second part of the century and at the higher
GHG prices.
GHG Mitigation by Individual Forestry
and Agricultural Activities: Annualized
Results
One way to summarize the net effects of
the differing time dynamics is to determine
a single measure of GHG effects over the entire
simulation period 2010 to 2110. The measure
employed here computes the annualized
equivalent GHG quantity effect. By annualizing
the estimates, one focuses more on comparing
mitigation quantities across activities and regions
and focuses less on comparisons across points in
time. Box 4-5 describes how the annualization
approach is applied to generate GHG mitigation
estimates in this study.
2 The analysis does not explicitly consider that avoiding CO2
emissions from fossil fuel might also have some elements of
impermanence as well. Avoided fossil fuel use simply retains
the carbon stock below ground for possible release in the
future. Although this is not as volatile and subject to rapid
release as terrestrial carbon, there are some risks of imper-
manence nonetheless. Non-CO2 emissions avoidance is
somewhat less prone to the impermanence effect than CO2
fossil fuel emissions.
Box 4-5: Annualizing Results over the Projection
Period
One way to summarize the net effects of the differing time
dynamics is to determine a single measure of GHG
effects over the entire simulation period 2010 to 2110. By
annualizing the estimates, one can focus more on broadly
comparing mitigation quantities across scenarios,
activities, and regions and focus less on comparisons
across specific points in time.
The annualized value provides a single measure that
essentially "smoothes out" variability over time, while
using the notion of time discounting to enhance the value
of near-term mitigation over mitigation occurring in the
distant future. Herzog et al. (2003) discuss the rationale
for using time-discounting concepts to quantify physical
mitigation quantities over time. Note that the annualiza-
tion approach outlined here is appropriate only when
GHG prices are constant over time. Therefore, only the
constant-price scenarios in this report are reported using
annualized estimates.
The annualized measure is computed by first taking the
net present value of the GHG mitigation quantities over
T
time: NPVG = £ Gt/(1+r)1, where G, is the GHG effect in
1=1
time period (decade) t; T is the length of the simulation
(in this case 100 years); and r is the annual discount rate,
which is 4 percent for this analysis. The NPVG value in the
equation above is then annualized via the following
calculation: GA = NPVG * AF, where AF is the annualization
factor for converting a lump sum present value, such
as NPVG into its annualized equivalent. For a 100-year
time period evaluated at a 4 percent discount rate, the
AF is 0.0408. The formula for the annualization factor
isAF = r(1+r)T/[(1+r)T-1].
Figure 4-7 shows the effect of providing a single annual-
ized value for a highly variable time trend such as the
annual mitigation estimates for the $15/t CO2 Eq. price
scenario. In the figure, the actual projected annual
values vary from about +900 Tg CO2 Eq. per year in the
middle of the projection to -300 Tg CO2 Eq. per year
toward the end of the projection, reflecting the carbon
reversal pattern discussed earlier in this chapter. The
annualized mitigation quantity using the formula
referenced above is 667 Tg CO2 Eq. per year (the flat
horizontal line in Figure 4-7). The annualized line can
be compared to the third line in Figure 4-7, which is the
cumulative annual average over the entire projection
period from 2010 to the point in time referenced in the
figure. Note that the three annual values (actual, cumula-
tive average, and annualized) are fairly close in value
for the first several decades of the projection. Then, as
carbon reversal occurs, the actual annual values drop
sharply and the cumulative annual estimate drops
gradually, while the annualized value, by definition,
stays fixed.
4-10
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Box 4-5: (continued)
Figure 4 7: Comparison of Actual, Cumulative
Average, and Annualized GHG
Mitigation Value Calculations at
$15/tCO2Eq.:2010 2110
(5
I
U'
HI
CM
O
o
O)
2,000 -i
1,000-
500-
0-
-500-
-1,000-
-1,500-
-2,000 -
20
nSJL _—
* %^~^~*-*7i
^^/
10 2030 2050 2070 2090 2110
Year
— A — Annualized
— ^—Actual Annual
— •— Cumulative Annual Average
The FASOMGHG model allows projection of scenarios
out for 100 years; however, policy time frames are likely
to be shorter than that. Indeed the results discussions
above have tended to focus on results for the first 40 to
50 years after the mitigation scenario is initiated. This
raises the question of whether results should be annual-
ized over time frames shorter than 100 years. The results
in Figure 4-7 suggest this could make a difference in
quantifying a scenario's GHG benefits. To demonstrate
this point, Table 4-4 shows how shortening the time
horizon for quantifying GHG effects from 100 years to 50
years and 20 years, respectively, changes the annualized
mitigation quantity estimate.
The first column in the table presents annualized quantity
estimates for each activity and all activities combined
when all projected values over the 100-year projection
period (positive and negative) are applied to the annual-
ization formula above. As shown in Figure 4-7, the total
quantity is about 667 Tg CO2 Eq. per year. When the
annualization is performed over a 50-year period, all
effects after 2060 are ignored. This produces a larger
annualized estimate (about 760 Tg) because the future
reversal of forest and soil carbon in the latter half of the
century is not deducted. Shortening the time horizon to
20 years increases the annualized estimate even further
(about 790 Tg), because none of the carbon reversal from
afforestation and soil carbon management is included
(some was included in the 50-year estimate) and thus
only the positive accumulations are taken into account.
One factor, though, that diminishes the 20-year estimate
relative to the 50-year and 100-year estimates is that the
latter two include biofuels, and the first estimate does
not. The reason that the 20-year estimate does not
include biofuels is that biofuel demand will not be
sufficient to induce production for several decades
at this price ($15/tonne) under assumptions maintained
in this analysis. The sensitivity of the model results to
the biofuel demand assumptions is explored later in
this chapter.
In summary, time dynamics are an important part of
the GHG mitigation story in forestry and agriculture,
and these effects are emphasized in a number of places
throughout this report. However, an annualized estimate
provides a theoretically consistent approach to capture
these dynamic GHG effects in a single measure, thereby
allowing for broad comparisons of mitigation quantities
across activities, regions, and price scenarios. The
annualized estimate depends on the length of time over
which the GHG effects are considered (e.g., 20, 50, ...
100 years). For the purposes of this report, the annual-
ized estimates will typically be presented for the 100-
year time horizon, because this is the most complete
estimate available and does not ignore potentially
important reversal effects in the distant future.
Table 4-4: Comparison of Annualized GHG Mitigation Estimates (Tg CO2 Eq. per year) across
Alternative Time Horizons at a GHG Price of $15/t CO2 Eq.
Annualized over...
Activity
100 Years
50 Years
20 Years
Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All Strategies
137.3
219.1
168.0
53.0
32.0
57.2
666.7
164.5
258.7
190.0
46.3
34.5
65.1
759.1
220.0
244.7
243.9
41.6
38.2
0.0
788.4
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-11
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Table 4-5 presents the annualized GHG quantity
effects for each major mitigation option by each
constant-price scenario. These data constitute a
GHG mitigation supply function for U.S. forestry
and agriculture, as illustrated in Figure 4-8. The
table and figure show that agricultural soil carbon
sequestration and forest management are the
dominant strategies at low prices, afforestation and
biofuels dominate at higher prices, and non-CO2
gas mitigation in agriculture plays a relatively
small role in sector strategies.
Annualized GHG Mitigation by Option.
Afforestation starts to take hold at the middle
price ($15) and becomes the dominant mitigation
strategy at the highest prices considered ($30 and
$50).3 This reflects higher opportunity costs of
converting agricultural land to forestland than for
changes in carbon management practices on
forestland and agriculture. It also demonstrates
that, once adopted, afforestation can have a larger
GHG impact than changes in management within
existing uses. Though, as shown above, these
effects are quite uneven over time.
Table 4-5: National GHG Mitigation Totals by Activity: Annualized Averages, 2010-2110
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline, annualized over the time
period 2010-2110.
Constant Prices Over Time
Activity
Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All Activities
$1
0.0
24.8
62.0
20.5
9.4
0.0
116.8
$5
2.3
105.1
122.7
31.9
15.2
0.1
277.3
$15
137.3
219.1
168.0
53.1
32.0
57.2
666.7
$30
434.8
314.2
162.4
77.6
66.8
374.6
1,430.4
$50
823.2
384.8
130.6
95.7
110.2
560.9
2,105.4
Figure 4 8: GHG Mitigation Supply Function from National GHG Mitigation Totals by Activity
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline, annualized over the time
period 2010 2110.
Afforestation
Forest management
Agricultural soil carbon
sequestration
Fossil fuel mitigation from
crop production
Agricultural CH4 and N2O
mitigation
Biofuel offsets
All activities
0 500 1,000 1,500 2,000 2,500
Emission Reduction in Tg CO2 Equivalent
3 The dominance of afforestation as a strategy is tempered somewhat by exogenous restrictions put on the aggregate contribution
of biof uel offsets from the forest and agriculture sectors to reflect current projections of potential biofuel demand by the United
States (Haq 2002). The effects of relaxing these biofuel demand restrictions are considered below.
4-12
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Forest management produces results much like
afforestation: fairly small amounts of GHG are
sequestered at the lower prices, and larger
amounts are only realized at the higher prices.
Although the amount of GHG mitigation at the
lower prices is small, forest management is second
only to agricultural soil carbon in terms of mitiga-
tion potential at the two lowest prices.
Agricultural soil carbon sequestration and forest
management are the dominant strategies at the
lower end of the GHG price range ($1 and $5 per
t CO2). This reflects the relatively low opportunity
cost associated with adopting reduced tillage or
altering forest management practices to sequester
more carbon in some places within the country.
These actions can produce results fairly early on.
The increase in other mitigation opportunities
actually leads to a slight decline in mitigation
through agricultural soil carbon sequestration
when moving from the $30 to $50 GHG price. This
is because land is being bid away from cropland at
these higher GHG prices; therefore, the land base
on which to modify tillage practice declines.
Fossil fuel mitigation in crop production plays
a very small role in total GHG mitigation at the
lower prices, increasing contributions at the higher
prices. However, even at the highest price scenario,
this activity accounts for less than 3 percent of
total mitigation in the first 2 decades.
Agricultural CH4 and N2O mitigation. Agricultural
non-CO2 gases are a substantial contributor to the
agricultural-sector baseline GHG emissions, as
shown in Chapters 2 and 3. However, the non-CO2
mitigation options provide somewhat limited
mitigation potential relative to the CO2 mitigation
and sequestration options.
The activities associated with non-CO2 gas reduc-
tions, such as enteric fermentation, manure man-
agement, and soil management, make their largest
relative contribution to aggregate mitigation at the
lowest price evaluated ($1), where they account for
8 percent of the mitigation portfolio. The share
drops to about 5 percent of the portfolio at the $5
price and remains at about 5 percent of total
mitigation for all prices above that.
One reason that mitigation potential for the non-
CO2 options is so limited in aggregate terms may
be the limited amount of data and other informa-
tion known about the biophysical and economic
consequences of these mitigation options (DeAn-
gelo et al. in press). Another factor may be that
what is known about some of the non-CO2 mitiga-
tion options shows that they are profitable under
BAU conditions and are thereby incorporated into
baseline practices, leaving fewer options available
for mitigation beyond the baseline. In either case,
more data and research may be needed to better
gauge the opportunities for non-CO2 mitigation
options in agriculture.
Biofuels are projected to play a substantially larger
role in the mitigation portfolio at higher GHG
prices and in later decades. Biofuel results are
predicted to increase more than tenfold from 2025
to 2055 (see Table 4.A.1 in the appendix).
Several factors contribute to the incidence and
timing of biofuel's role in the mitigation portfolio.
First, biofuels are largely uneconomic in the
baseline and would take a subsidy to become
economically competitive with other fuel sources.
A GHG price can serve, essentially, as such a
subsidy. As the incentive grows, so does biofuel
production. But as explained in Chapter 3, the
FASOMGHG model imposes exogenous limits on
biofuel demand capacity for several decades. As
these limits become less binding over time, adop-
tion increases significantly as well.
Biofuels also do not possess the same reversibility
effects as its main competing activities at the high
GHG prices. Whereas afforested lands are shown
to revert back to agriculture after several decades,
biofuel effects are more permanent, both in terms
of their ability to offset fossil fuel emissions in the
first place and their avoidance of future releases
of stored carbon through land-use change or
practice reversion.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-13
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Sensitivity of National-Level Results to Two
Key Assumptions. As discussed in Chapter 3,
the FASOMGHG model depends on a wide range
of data, parameters, and other assumptions that
determine the validity of the model simulations.
Of these factors, two stand out as particularly
worthy of further scrutiny: (1) the assumed time
it takes for a change in agricultural soil tillage
practices to achieve a new soil carbon equilibrium
(i.e., achieve its "saturation" point) and (2) the
assumed rate of market penetration for biofuel
demand. Boxes 4-6 and 4-7 present a sensitivity
analysis of FASOMGHG model results to changes
in these assumptions and finds that the national-
level results by activity are moderately affected by
changes in the assumed time to achieve the new
agricultural soil carbon equilibrium point and the
time profile of biofuel demand.
Box 4-6: Sensitivity Analysis of Key Assumption: Time to Reach Soil Carbon Equilibrium ("Saturation")
The FASOMGHG model results for agricultural soil
carbon sequestration could depend critically on the
assumed time period for soil carbon to reequilibrate to a
steady state (or "saturate" as described above) following
a change in tillage practice. In FASOMGHG, the annual
soil increment following a change in tillage practices is
calculated as follows:
ACt = (CSSR - CSSC)/TS
[4.1]
where AC, is the estimated annual change in year t; CSSR
and Cssc are the soil carbon steady-state values under
reduced tillage and conventional tillage, respectively;
and Ts is the time to steady state (equilibrium). The
carbon steady-state values are given by simulations of
the CENTURY model (Parton 1996), but CENTURY does
not simulate the Ts variable. Therefore, an assumed value
for Ts is needed. Note that ACt goes to zero once the
new steady state is reached. Therefore, both the size and
timing of the annual carbon increment are affected by
the assumed length of time to reach the new equilibrium.
The maintained assumption for the model simulations
thus far is that the soil carbon saturation period is 15
years, based on work by West and Post (2002). They
quantitatively synthesized the published results of 276
paired treatments of changes in tillage practices from 67
study sites and estimated that the new soil carbon
steady state was reached in 10 to 15 years. However,
other research has suggested possibly longer saturation
periods for tillage change (Lai et al. 1998). To evaluate
the sensitivity of the foregoing results to this assumption,
the FASOMGHG model was run with an assumed time
to equilibrium of 30 years and compared to the results
with the 15-year saturation period.
The simulation was run for a constant GHG price of $15,
which was selected because all of the mitigation activities
come into play at that price. The results in Figure 4-9 are
annualized national mitigation estimates for the projec-
tion period 2010 to 2110. The annualized contribution of
the agricultural soil carbon mitigation declines by almost
half, from about 170 Tg CO2 per year to 90 Tg per year,
which is about what one might expect when the time to
equilibrium is doubled, and therefore the annual incre-
ment calculation in equation [4.1] is halved (assuming
the same quantity of mitigation). However, that is not the
end of the story. The figure illustrates that not only is
there the expected reduction in annual mitigation from
agricultural soil carbon sequestration when the satura-
tion period is elongated, but also the contribution of
other activities is affected as well. In particular, the
reduction in agricultural soil carbon mitigation is partly
offset by increased mitigation from biofuel offsets and
agricultural CH4 and N2O mitigation and to a lesser
extent forest carbon and fossil fuel mitigation. The net
reduction in mitigation across all activities is under 50
Tg CO2 per year, so the initial 80 Tg reduction from soil
carbon is offset by about a 30 Tg net increase in the
other activities. In essence, this shows that GHG mitiga-
tion options compete with each other on a fixed land
base. When one option becomes less advantageous,
the competing options can take up some of the slack.
4-14
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Box 4-6: (continued)
Figure 4 9: Model Sensitivity to Saturation Period toward a New Soil Carbon Equilibrium from
Tillage Change: GHG Price = $15/t CO2 Eq.
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline, annualized over the time
period 2010 2110.
800
115-year
130-year
Afforestation Forest Agricultural Fossil fuel Agricultural Biofuel
management soil carbon mitigation CFU and N2O offsets
sequestration from crop mitigation
production
Activities
All
Strategies
Box 4-7: Sensitivity Analysis of Key Assumption: Biofuel Demand
The FASOMGHG model was modified in this report to
confine biofuel production to fall within the capacity
limits projected by the ElA's energy forecasts (Haq
2002). As such, some biofuel mitigation that may initially
seem profitable within FASOMGHG is excluded for
consistency with the EIA estimates. To test for the
sensitivity of this assumption, the model was re-run
to relax the EIA demand assumption and rely purely
on the profitability of biofuel production as a determi-
nant of total biofuels supplied to the market.
The results of this simulation are illustrated in Figure
4-10. The simulation was run at a GHG price of $30/t
CO2 Eq. (constant), which is the price at which biofuels
become a substantial contributor to national mitigation
totals. Relaxing the biofuel demand restriction raises the
contribution of that activity for sensitivity analysis from
375 to 530 Tg CO2 Eq. per year, more than a 40 percent
increase. As with the agricultural soil carbon example,
we must consider offsetting effects from the other
activities, but they are not all negative. The contribution
of afforestation declines as part of the mitigation
portfolio, but the contribution of agricultural soil carbon
and non-CO2 mitigation rises, indicating there are
complementarities between biofuel production and
mitigation from these activities. Notably, land that is
diverted from traditional crops to biofuel production
tends to sequester more carbon and release less N2O
and CH4.
Figure 4 10: Sensitivity of Model Results to Assumed Biofuel Demand Restrictions:
GHG Price = $30/t CO2 Eq.
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline, annualized over the time
period 2010 2110.
1,800
i Restricted
Demand
I Unrestricted
Demand
Afforestation Forest Agricultural
management soil carbon
sequestration
Fossil fuel
mitigation
from crop
production
Agricultural
CH4 and N2O
mitigation
Biofuel
offsets
All
Strategies
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-15
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
GHG Mitigation by Region
Because the U.S. landscape is quite heterogeneous,
the adoption and effectiveness of GHG-mitigating
activities will not be uniform across regions within
the country. The regional definitions used in this
section can be found in Table 3-2 in Chapter 3.
The regional totals distribution at the middle three
constant-price scenarios ($5, $15, and $30/t CO2
Eq.) are illustrated in Figure 4-11. This figure and
the corresponding table (Table 4.A.2 in the appen-
dix) with activity detail provide a summary of
annualized GHG mitigation quantities by major
region, activity, and price scenario. Table 4.A. 3
in the appendix reports the regional breakdown
of annualized mitigation totals by all key activities
modeled.
By and large, the regions with the highest GHG
mitigation are the South-Central, Corn Belt, and
Southeast regions. At the lower GHG prices, the
Lake States and Great Plains are key contributors
as well. The contributions of the Corn Belt, Lake
States, and Great Plains are primarily in the form
of agricultural soil carbon sequestration, whereas
the South-Central and Southeast regions are
primarily suppliers of carbon sequestration from
afforestation and forest management.
The Rockies, Southwest, and Pacific coast states
generate relatively small shares of the national
mitigation total under all of the price scenarios.
From those regions, only forest management from
the PNWW produces appreciable mitigation. This
is because climate and topography significantly
limit the movement of land between major uses
such as forestry and agriculture in the western
regions.
When biofuel production is selected at the higher
GHG prices, this occurs primarily in the North-
east, South, Corn Belt, and Lake States.
Figure 411: Total Forest and Agriculture GHG Mitigation by Region
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline, annualized over the time
period 2010 2110.
500
NE
SE
LS
CB
SC
GP
SW
RM
PNWE
$30
$15 GHG Price
$5 ($/tC02Eq.)
PNWW
Region
PSW
4-16
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Table 4-6 presents a top 10 ranking of region-
activity combinations producing the most GHG
mitigation by price scenario. This table illustrates
how the distribution of GHG mitigation opportu-
nities varies across regions and activities as the
GHG price changes. At the lowest two prices, the
top-ranked combination is forest management in
the South-Central region, followed by agricultural
soil carbon sequestration in the Corn Belt and
Lake States. As prices rise, so do the opportunities
for afforestation in the South-Central and Corn
Belt regions and biofuel production in the Corn
Belt, South, and Northeast.
Non-GHG Environmental Co-effects
The undertaking of GHG mitigation activities and
the resultant shift of land uses and management
practices have the potential to produce environ-
mental co-effects other than climate change mitiga-
tion. For instance, the changes in agricultural prac-
tices can have an effect on the farm inputs applied,
which in turn can affect the loadings of nutrients,
erosion, and other residuals into waterbodies.
Table 4-6: Top 10 Region-Activity Mitigation Combinations
Ranks are based on mitigation quantities annualized over the period 2010-2110.
GHG Constant Price Scenario ($/t CO2 Eq.)
Region
sc
CB
LS
GP
SW
RM
SC
NE
CB
CB
SE
SC
NE
RM
SW
CB
SE
SC
CB
LS
Activities $1
Forest management 1
Agricultural soil carbon sequestration 2
Agricultural soil carbon sequestration 3
Agricultural soil carbon sequestration 4
Fossil fuel mitigation from crop production 5
Agricultural soil carbon sequestration 6
Fossil fuel mitigation from crop production 7
Agricultural soil carbon sequestration 8
Fossil fuel mitigation from crop production 9
Agricultural CH4 and N2O mitigation 10
Forest management
Afforestation
Biofuel offsets
Afforestation
Agricultural soil carbon sequestration
Afforestation
Biofuel offsets
Biofuel offsets
Biofuel offsets
Afforestation
$5 $15 $30 $50
1133
2 4 7 10
3 6
5 7
7
8
6 8 10
9
10
4368
2 1 2
545
9
10
2 1
5 4
8 6
9 7
9
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-17
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
To briefly assess these effects, the analysis focuses
on a single GHG price ($15/t CO2 Eq.), as shown
in Figure 4-12. Three of the four pollutants reveal
a reduction in overall loadings relative to baseline
amounts. Phosphorous and erosion loadings reveal
the largest reduction of approximately 40 percent
each. This reduction in pollutant loadings is tied to
the widespread adoption of conservation or zero
tillage practices, which reduces erosion and
phosphorous runoff that often adheres to soil
particles.4 Over time, however, these loadings
return closer to baseline levels. Pesticides are
the only loadings that exceed baseline loadings
in some cases. This finding reflects the fact that
adopting no-till farming practices often requires
increased pesticide applications, as chemical
means of weed control replace mechanical means.
Chapter 7 expands the discussion of environ-
mental co-benefits by evaluating the full range
of constant GHG prices, evaluating the net likely
impact of these loadings patterns on water quality
and considering other environmental co-effects
such as biodiversity.
Mitigation Response to Rising GHG Price
Scenarios
Up to this point, the chapter has focused on results
for the constant GHG price scenarios. Now results
from the rising-price scenarios are discussed. The
focus of the discussion is primarily on the differ-
ences from the constant-price results. A detailed
table of mitigation results by activity in key years
for the rising-price scenarios is presented in the
appendix to this chapter (Table 4.A.4).
As with the constant-price scenarios, there is a
larger amount of GHG mitigation with the higher
rising-price scenarios; however, the major differ-
ence between the constant- and rising-price
scenarios is the timing of the mitigation. These
timing effects are illustrated in Figure 4-13. As
shown earlier, the GHG mitigation totals start
high in 2015 and then decline by 2055 under
the constant-price scenarios. The rising-price
scenarios, however, tend to show the opposite
effect. Mitigation is minimal in the early years
when prices are low but rises substantially in the
later years as the prices escalate for two of the
Figure 4 12: Pollutant Loading Effects Over Time of a $15/t CO2 Eq. GHG Price
120
100
60
40
2010 2020 2030 2040
Year
Note: All values indexed to a baseline value of 100.
'Baseline
• Nitrogen
Phosphorous
• Erosion
• Pesticides
2050
2060
4 Recall from Table 4-3 that the $15 carbon price in the year 2015 resulted in the largest conversion of conventional till to either
conservation or zero tillage practices.
4-18
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
three scenarios. To a large extent, this time pattern
of mitigation is the result of the producers of GHG
mitigation holding out for the higher prices that
occur in the later years of the projection. This
is particularly crucial with mitigation options
because carbon sequestered early on cannot be
re-sequestered in the future. When prices are
expected to rise, this provides an incentive to wait
on enacting sequestration activity.
Figure 4-14 illustrates cumulative GHG effects
over time for the two scenarios that have an initial
price of $3 and rise at 1.5 percent and 4 percent,
respectively. The main differences between the
two scenarios are as follows:
• The scenario with the 4 percent rate of increase
demonstrates a substantial delay in mitigation
activity, as suppliers wait for the much higher
prices to come in the future. Once prices near
their $30 cap at mid-century, significant action
takes hold.
• The level of mitigation ultimately obtained is
substantially larger in the 4 percent scenario,
primarily because the price gets much higher
in the out years. As such, the biofuel option
becomes more attractive. The biofuel option also
favors later adoption because the demand for
biofuels over time reflects the assumption that
the capacity for biofuel use in electricity genera-
tion is heavily constrained in the short run but
could expand substantially in the long run.
Figure 4-15 shows cumulative GHG mitigation for
the more aggressive rising-price scenario, starting
at $20It CO2 Eq. and rising to $75. This case also
produces delay in mitigation but includes a much
larger quantity of mitigation than the other two
scenarios and has a larger role for afforestation
because of the higher starting and ending prices.
These figures reveal the expected differences
resulting from the higher prices, while highlight-
ing the timing effects that are not seen in the
constant-price scenarios.
Figure 4 13: Constant Price Scenarios vs. Rising Price Scenarios and GHG Mitigation
Quantities are Tg CO Eq. per year net emissions reduction below baseline for 2015 and 2055.
2,500
2,000 -
iS" 1,500 -
CNJ
o
o
en 1,000 -
500-
0
JZL
IL
T
J
$1
$5 $15 $30 $50
i r
$3 @ $3 @ $20 @
1.5% 4% $1.30
2015
2055
Note: All values indexed to a baseline value of 100.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-19
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Figure 4 14: Cumulative GHG Mitigation over Time: $3/t CO2 Price Rising at Two Rates
Quantities areTg CO2 Eq. cumulative net emissions reduction below baseline.
$15/t CO Eq. Eq. Price Rising at 1.5% per Year
O
O
01
O
O
O)
D Biofuel offsets
D Crop management FF mitigation
D Ag CH4 and N2O mitigation
• Forest management
D Afforestation
• Agricultural soil C sequestration
70,000
60,000
50,000 -
40,000
30,000 -
20,000
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
$30/t CO Eq. Rising at 4% per Year ($30 cap)
10,000-
D Biofuel offsets
D Crop management FF mitigation
• Ag CH4 and N2O mitigation
• Forest management
D Afforestation
• Agricultural soil C sequestration
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
Figure 4 15: Cumulative GHG Mitigation over Time: $20/t CO2 Price Rising by $1.30 per Year ($75 cap)
Quantities areTg CO2 Eq. cumulative net emissions reduction below baseline.
D Biofuel offsets
D Crop management FF mitigation
D Ag CH4 and N2O mitigation
• Forest management
D Afforestation
• Agricultural soil C sequestration
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
4-20
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Comparison of FASOMGHG Results
with Other Analyses
It is useful to compare the results of the analysis
presented in this chapter to similar economic
studies of GHG mitigation in the U.S. forest and
agriculture sectors. It is important to note, however,
that this study is rather unique in terms of its
depth and breadth of mitigation options covered
across the two sectors. In essence, this is a some-
what more comprehensive and integrated assess-
ment of economic potential of the U.S. forest and
agriculture sectors together than other studies
to date. So a direct and consistent comparison
with other studies is not quite possible. However,
several studies have looked separately at the
national mitigation potential from afforestation,
forest management, and agriculture and can
thereby provide context for the core results
presented above.
Richards and Stokes (2004): Forest Carbon
Richards and Stokes (2004) conducted a thorough
review of 36 forest carbon sequestration economic
studies throughout the world. Among this group,
eight studies estimated marginal cost functions for
forest carbon sequestration at the national level for
the United States, reportable on an annual basis.
Consequently, these eight studies are directly
comparable to the results presented in this chapter,
once the appropriate adjustments are made to
tonnes of CO2 Eq. per year.5 Table 4-7 summarizes
the range of carbon sequestration quantity and
cost results for the eight comparable U.S. studies
reviewed by Richards and Stokes and compares
them to the results from the constant-price
FASOMGHG simulations in this study. The
aggregate national forest carbon sequestration
estimates in the Richards and Stokes studies
ranged from 147 to 2,349 Tg CO2 Eq./yr at a cost
(price) ranging from $1.36 to $40.87 per t CO2 Eq.
Most of these studies examine afforestation only
or do not break out afforestation from forest
management. Only one of the studies presents
results for forest management activities, and that
study produced an estimate of roughly 400 Tg
CO2 Eq./yr of sequestration at a cost ranging from
$1.63tol2.81/tCO2Eq.
Many compounding factors cause the results to
vary widely in the studies reviewed by Richards
and Stokes, including but not limited to the extent
of ecosystem components included in the carbon
calculations, the biophysical foundation for the
Table 4-7: Comparison of FASOMGHG Results in this Chapter to Range of Estimates from Richards
and Stokes' (2004) Review Study
Carbon Sequestration (Tg CO 2 Eq. per Year)
This Study:
Comprehensive Activities,
Annualized Over 2010-2110
GHG Price Scenario ($/t CO2 Eq.)
Richard and Stokes:
U.S.-Based Studies
GHG Price Range ($/t CO2 Eq.]
Activity
$5
$15
$30
$50
$1.36 - $40.87
Afforestation 2.3 137
Forest management 105 219
Total forest carbon 107 356
435
314
749
823
385
1,208
147-2,349
404a
551 -2,753
Only one study covering the United States included estimates for forest management.
5 The eight comparable studies are Moulton and Richards (1990), Adams et al. (1993), Parks and Hardie (1995), Callaway and
McCarl (1996), Alig et al. (1997), Richards (1997), Adams et al. (1999), and Stavins (1999). Unfortunately, Richards and Stokes did
not adjust the studies' results to put them in a common year for dollar comparisons.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-21
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
carbon sequestration rates used, and the land costs
included in cost calculations. However, comparing
the U.S. forest carbon sequestration estimates
generated by the FASOMGHG results earlier in the
chapter suggests they fall well within the range of
estimates found in the Richards and Stokes review.
FASOMGHG mitigation estimates will generally
not reach the high end of the estimates found in the
Richards and Stokes study, because FASOMGHG
employs economic feedback effects (e.g., timber
and agricultural price effects) that will temper
sequestration responses, in contrast to studies that
estimate mitigation cost functions without market
feedback effects.
Stavins (1999): Afforestation
For a further comparison of this chapter's results
to other studies, we look at research conducted by
Stavins (1999) that synthesized the results from
several past studies that were directly comparable
to the results presented in his work in that they
were national (United States) in scale and focused
specifically on afforestation. Stavins computes a
95 percent confidence interval on his national
marginal cost function for afforestation and shows
that other previously published studies (Richards
et al. 1993, Adams et al. 1993, and Callaway and
McCarl 1996) fall within that interval.
To compare the results from this study to Stavins',
several adjustments needed to be made. First,
Stavins' results are presented graphically via a
marginal cost function. This enabled one to trace
the amount of carbon sequestered nationally to a
given level of marginal cost per tonne sequestered.
Conceptually, this is similar to evaluating the total
amount of carbon that can be sequestered at a
given GHG price. This enables direct comparison
with the FASOMGHG results presented above.
However, further adjustment is necessary to
compare Stavins' results, which are expressed in
short tons of carbon and 1990 dollars, with the
results here, which are in tonnes of CO2 equivalent
and 2000 dollars.6 These adjustments are made and
results are compared in Table 4-8 for the $30
and $50 constant-price scenarios, which are the
two scenarios in which forest carbon plays the
largest role.
The main implication from the comparative
results presented in Table 4-8 is that the core
scenario analysis in this report suggests a smaller
aggregate potential for forest carbon sequestration
than that found in the Stavins study. When this
study's afforestation carbon potential is compared
to Stavins, which is the most relevant comparison,
the mitigation quantities are about one-third
to one-half of Stavins' estimates. When forest
management is added to the totals from this
study, the relative quantities are one-half to
three-quarters of the Stavins' estimates.
Table 4-8: Comparison of FASOMGHG Results in
this Chapter to Stavins' (1999) Study
Carbon Sequestration
(Tg CO2 Eq. per Year,
above baseline,
annualized over
100-year time period)
ion
"
GHG Price ($/t CO2 Eq.)
$30
$50
This Study
Afforestation
Forest management
Total forest carbon
435
314
749
823
385
1,208
Stavins' Central Estimate3 1,330 1,660
This Study as % of Stavins'
Afforestation 33% 49%
Total forest carbon 56% 73%
Adjustments made to convert Stavins' estimates from
1990 dollars per short ton to 2000 dollars per t CO2 Eq.
6 Short tons of carbon are converted to tonnes by dividing by 1.102. Tonnes of carbon are converted to tonnes CO2 by multiplying
by 3.667.1990 dollars are converted to 2000 dollars using the consumer price index (urban consumers)
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Stavins' paper asserts that one might typically
expect econometric estimates, like those in his
study, to yield smaller mitigation quantities than
estimates using optimization methods like the
FASOMGHG model, because of the econometric
reliance on "revealed preferences" of landowners.
However, while FASOMGHG does not incorporate
the revealed behavior of an econometric model, it
does capture (unlike the Stavins study) feedbacks
from the commodity and land markets that need
to be considered when estimating the net effects
of large-scale programs. Large-scale movement of
land from agriculture to forests will tend to raise
agricultural prices and lower timber prices. This
provides an incentive for countervailing move-
ments of land from forest to agricultural use. The
multimarket equilibrium nature of FASOMGHG
captures these feedbacks and slows the afforestation
(and sequestration) process accordingly. Ignoring
this feedback tends to overstate sequestration
potential all else equal, as Stavins acknowledges
in his paper.
Sedjo, Sohngen, and Mendelsohn (2001):
Forest Carbon
Since the Stavins (1999) study, other forest carbon
sequestration studies have been published that are
in some ways comparable to those synthesized by
Stavins (see, for instance, Adams et al. [1999],
Plantinga et al. [1999], Stavins and Newell [2000],
Sedjo, Sohngen, and Mendelsohn [SSM] [2001],
and Sohngen and Mendelsohn [2003]). Perhaps
the most directly comparable of those studies is
the SSM 2001 study, which looks at a wide range
of price scenarios similar to the constant-price
scenarios in this chapter. The one important
difference, though, is the SSM results are for all
of North America, while these results are for the
United States. Nevertheless, U.S. results are by
far the dominant component of the North America
results in SSM. Table 4-9 compares SSM results at
$50 and $100 per tonne of carbon ($13.62 and $27.25
per t CO2 Eq.) with the closest points of compari-
son in this study ($15 and $30 per t CO2 Eq.).7
The SSM mitigation estimates are about one-
quarter less than the FASOMGHG results under
both price levels. While this is somewhat surpris-
ing given the larger continental coverage of the
SSM study, many modelers would consider a 25
percent variation in such macro-scale results using
two different models a reasonably good correspon-
dence. Further examination of the two models'
results suggests that the differences are primarily
due to the more detailed modeling of land oppor-
tunity costs in U.S. agriculture in FASOMGHG.
This produces a more elastic afforestation
response than the SSM study, which relies on
a single inelastic land-use supply function from
agriculture.
Table 4-9: Comparison of FASOMGHG Forest Carbon Sequestration Results in this Chapter with Sedjo,
Sohngen, and Mendelsohn (2001)
Quantities for both studies areTg CO2 Eq. per year, sequestration above baseline, annualized over
100-year time period.
Sedjo, Sohngen, and
Mendelsohn (2001)
Scenario
$13.62/tCO2Eq.
($50.00/t C Eq.)
$27.25/t CO2 Eq.
($100/tCEq.)
Total Forest Carbon
Sequestration
(Tg CO2 Eq. per Year)
265
563
This Study
Scenario
$15/tCO2Eq.
$30/t CO2 Eq.
Total Forest Carbon
Sequestration
(Tg CO2 Eq. per Year)
356
749
7 The direct comparison between this study's results and those of SSM was enabled with data provided by Dr. Sohngen that is not
directly presented in one of the paper's tables.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-23
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
USDA, Economic Research Service (2004):
Agricultural Carbon Sequestration
Most recently a report by the USDA ERS
was published that examined the economics of
sequestering carbon in the agriculture sector
(Lewandrowski et al. 2004). That report examines
mitigation options in the agriculture sector, includ-
ing afforestation but excluding forest management
and biofuels. The ERS study produced estimates
for the amount of carbon that could be sequestered
over a 15-year time period given various carbon
prices expressed in $/t C. After converting these
to $/t CO2 Eq. the prices range from $2.72 to $34.05
per tonne (see Table 4-10).
These prices are introduced in a model of the
U.S. agriculture sector (USMP), which is a spatial
market equilibrium model. All mitigation estimated
by this model is relative to a baseline generated
by the model. The USMP model results are also
separated by forest and soil sequestration, allow-
ing for a comparison to the FASOMGHG soil
results. At the lowest GHG price, the amount
of overall carbon sequestered ranged from 0.4
to 35 Tg CO2 Eq. per year. The highest price
investigated resulted in total sequestration
ranging from 237 to 587 Tg CO2 Eq. per year.
The range of estimates presented in the USDA
ERS report is generally lower than the range of
estimates generated by FASOMGHG in this study,
for a comparable set of activities and time horizon
(15 years). These differences can be expected
Table 4-10: Comparison of this Study with Lewandrowski et al. (2004) (USDA ERS)
based on the differences in the models and
assumptions embedded in the estimates. Note
that the FASOMGHG estimates for these price
scenarios are lower when we look over time
periods longer than 15 years. However, we cannot
compare longer time horizon estimates to the
ERS study, which takes a static snapshot of a
15-year program.
Recap of Study Comparisons
Although not a comprehensive comparison of
the results of this study to the entire spectrum
of results in the literature, the comparisons above
provide some validation that the results of various
components analyzed here are within the (fairly
wide) range of mitigation estimates found in
similar economic studies. Differences across the
studies can be explained in large part by differ-
ences in methodology and geographic coverage.
Taken together, these comparisons suggest that
the FASOMGHG model produces results that,
while more comprehensive in its coverage of both
forestry and agriculture than most other studies,
are consistent with findings on different compo-
nent parts (afforestation, forest management,
and agricultural soil carbon sequestration).
lis Study
(Tg CO2 Eq./yr net emissions
reduction below baseline)
After 15 years (Yr. 2025)
USDA ERS
(Tg C02 Eq./yr)
Average annual mitigation for 15-year program
GHG Price
($/t C02 Eq.)
Afforestation
Agricultural
soil carbon
sequestration
Total
$5
12
149
161
$15
228
204
432
$30
806
187
994
$50
1,296
153
1,449
$2.72
0-31
0.4-4
0.4-35
$6.80
20-140
3-10
25-151
$13.60
105-264
3-30
108-295
$20.40
145-378
5-48
151-426
$27.50
1 74-460
11-70
1 85-529
$34.05
224-489
13-95
237-587
4-24
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Appendix 4.A
This appendix provides detailed tabular results that
Table 4.A.1 : Key Results at the National Level by
are referenced in the
Activity, Time Period,
Quantities are Tg C02 Eq. per year net emissions
representative years 2015, 2025, and
2055.
reduction
main text
of this chapter.
and Constant-Price Scenarios
below baseline for
GHG Price ($/t CO
Year3
2015
2025
2055
Activity
Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All activities
Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All activities
Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All activities
$1
0
27
66
17
11
0
121
0
22
67
14
7
0
110
1
-10
1
14
7
0
13
$5
0
121
139
23
15
0
298
12
89
149
18
17
0
285
-7
48
-26
49
11
0
74
$15
145
227
194
35
28
0
629
228
156
204
32
36
0
655
-270
171
-22
62
26
121
86
2Eq.)
$30
557
271
191
46
48
16
1,129
806
250
187
49
76
21
1,390
-873
322
-10
92
52
990
572
$50
877
301
177
55
69
17
1,496
1,296
309
153
62
119
83
2,021
-426
325
-30
111
101
1,021
1,101
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-25
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Table 4.A.2: Total Forest and Agricultural GHG
Mitigation by Region
Quantities are Tg CO2 Eq. per year net
emissions reduction below baseline,
annualized over the time period
2010-2110.
Table 4.A.3: Forest and Agricultural GHG Mitiga-
tion by Activity, Region, and Price
Scenario
Quantities are Tg CO2 Eq. per year net
emissions reduction below baseline,
annualized over the time period 2010-2110.
Region
NE
SE
LS
CB
SC
GP
SW
RM
PNWE
PNWW
PSW
GHG Price ($/t CO2 Eq.)
$5 $15 $30 Region
10.9 64.7 148.1 Afforestation
36.4 92.6 236.0 CB
34.6 44.8 84.9
PNWE
49.0 80.8 326.4
PSW
83.9 278.1 507.5 RM
20.5 27.3 25.5 sc
18.1 26.7 31.7 SE
15.3 29.8 32.7 US
$5
2.0
0.0
0.3
0.0
0.0
0.0
0.0
2.3
GHG Price ($/t CO2
$15
6.6
0.0
1.6
1.6
11.7
115.8
0.0
137.3
Eq.)
$30
162.5
14.9
2.3
2.4
11.8
228.6
12.4
434.8
2-2 4-3 4-8 Forest Management
3.2 9.6 19.1 CB
3.2 8.0 13.8 LS
NE
PNWE
PNWW
PSW
RM
SC
SE
US
Agricultural Soil
CB
GP
LS
NE
PNWE
PSW
RM
SC
SE
SW
US
(continued)
-3.0
0.8
1.9
0.2
3.2
0.7
1.9
70.6
28.8
105.1
-5.6
5.7
9.5
0.2
9.6
0.8
2.0
127.7
69.2
219.1
-5.5
14.2
23.6
0.4
19.1
2.9
4.7
160.8
93.9
314.2
Carbon Sequestration
39.5
20.0
33.3
6.9
1.5
0.3
7.5
4.5
3.8
5.5
122.7
62.2
29.3
36.9
4.7
2.4
0.7
9.5
4.3
7.6
10.5
168.0
72.4
33.2
33.1
-3.7
2.7
0.9
9.6
-6.0
7.0
13.2
162.5
4-26
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Table 4.A.3: Forest and Agricultural GHG Mitiga-
tion by Activity, Region, and Price
Scenario (continued)
GHG Price ($/t CO2 Eq.)
Table 4.A.3: Forest and Agricultural GHG Mitiga-
tion by Activity, Region, and Price
Scenario (continued)
GHG Price ($/t CO2 Eq.)
Region
$5
$15
Fossil Fuel Mitigation from Crop Production
CB
GP
LS
NE
PNWE
PSW
RM
SC
SE
SW
US
Agricultural CH4
CB
GP
LS
NE
PNWE
PSW
RM
SC
SE
SW
US
Biofuel Offsets
CB
GP
LS
NE
PNWE
PSW
RM
SC
SE
SW
US
(continued)
6.5
1.0
0.4
1.1
0.2
1.3
1.2
10.2
1.3
8.7
31.9
and A/2
4.1
-0.8
0.1
0.9
0.0
0.9
4.7
-1.1
2.5
3.9
15.3
-0.1
0.3
0.1
0.0
0.0
0.0
0.0
-0.3
0.0
0.1
0.1
10.5
0.8
1.0
1.7
0.2
2.3
1.3
23.7
1.9
9.7
53.1
O Mitigation
7.4
-3.3
1.1
1.0
-0.1
2.7
5.2
6.9
4.7
6.4
32.0
-0.3
0.6
0.1
47.9
0.0
0.0
0.0
-0.4
9.2
0.1
57.2
$30 Region $5
All Activities
21.7 CB 49.0
-0.4 GP 20.5
1.8 LS 34.6
1.2 NE 10.9
0.0 PNWE 2.2
3.4 PNWW 3.2
1.4 PSW 3.2
33.4 RM 15.3
5.8 SC 83.9
9.3 SE 36.4
77.6 SW 18.1
US 277.3
24.2
-8.5
1.6
1.8
-0.6
4.3
5.1
21.0
9.2
8.9
66.8
51.1
1.1
19.3
125.1
0.1
0.0
0.2
69.9
107.5
0.3
374.6
$15 $30
80.8 326.4
27.3 25.5
44.8 84.9
64.7 148.1
4.3 4.8
9.6 19.1
8.0 13.8
29.8 32.7
278.1 507.5
92.6 236.0
26.7 31.7
666.7 1,430.4
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
4-27
-------
CHAPTER 4 • MITIGATION POTENTIAL: COMPREHENSIVE SCENARIOS WITH ALL ACTIVITIES AND ALL GHGS
Table 4.A.4: Key Results at the National Level by Activity, Time Period, and Rising Price Scenarios
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline for representative years
2015, 2025, and 2055.
Year" Activity
2015 Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All activities
2025 Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All activities
2055 Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All activities
$20@$1.30/yr
132
101
105
38
31
4
411
649
176
135
47
59
153
1,218
565
423
-26
113
101
1,021
2,196
$3@1.5%/yr
0
61
103
20
13
0
198
4
21
116
17
15
0
174
-3
19
-3
50
12
0
75
$3 @ 4%/yr
7
62
25
21
14
0
129
11
-67
48
18
18
0
28
15
141
76
62
25
352
671
a Year represents midpoint of decade tracked in FASOMGHG model (e.g.,2015 represents the midpoint of the 2010 to 2019
decade).
4-28 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 5
Mitigation Potential
of Selected Activities
Chapter 5 Summary
GHG mitigation for forestry and agriculture is considered on a more limited scale than the com-
prehensive coverage assessed in Chapter 4. Scenarios include fixed time-specific (Year 2025 and
Year 2055) GHG mitigation quantities from forestry and agriculture, payments for CO2 only (vs. for all
GHGs), and payments for selected mitigation activities.
For fixed time-specific scenarios, the effectiveness of GHG mitigation depends on the size of the
fixed mitigation quantity and whether efforts to maintain that level of mitigation remain in place or
expire. Aiming for future annual mitigation levels could lead to unintended GHG releases in preced-
ing years. This is particularly relevant for forest carbon. Aiming for cumulative, rather than annual,
mitigation could address this problem.
Paying for CO2 mitigation only does not significantly diminish the net GHG mitigation potential of
forestry and agriculture compared to scenarios where payments for all GHGs are made, since most
GHG mitigation occurs through sequestration and CO2 reductions. Non-CO2 reductions prove to be
complementary to—and thus occur with—CO2 mitigation.
Scenarios in which only agricultural activities are carried out can achieve moderate levels of
GHG mitigation, even at fairly low cost. Forest carbon sequestration and biofuels contribute more
substantially at somewhat higher price scenarios or when price scenarios rise over time. Agricultural
GHG mitigation opportunities are widely distributed across the United States, but most forest GHG
mitigation opportunities occur in the South.
The previous chapter evaluated GHG mitiga-
tion potential under scenarios for all three
critical GHGs (CO,,, CH4, and N2O) across
all agricultural activities and carbon sequestration
options in forestry and agriculture. As the results
indicate, a comprehensive payment approach has
the potential for large-scale mitigation, potentially
generating up to 2,000 Tg CO2 (2 billion tonnes CO2
Eq., or about 550 Tg C Eq.) per year of mitigation.
However, for several reasons forestry and agricul-
ture's role in national GHG mitigation might
involve less than comprehensive coverage of all
activities and GHGs (Sampson 2003; Richards et
al. forthcoming):
Much of the focus to date on GHG mitigation
has been on emissions from energy-producing
sectors, while the role of forestry and agriculture
has been seen more as a cost-effective means to
offset emissions from these other sectors.
Some GHG-emitting (sequestering) activities in
forestry and agriculture are difficult to measure,
monitor, and verify and could thereby be diffi-
cult to include in a comprehensive accounting
and incentive approach.
Individual sources of emissions and sequestra-
tion tend to be small and widely dispersed over
the landscape, making cost-effective aggrega-
tion of mitigation activities potentially difficult.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
5-1
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
Because of these issues, it is reasonable to evaluate
smaller-scale mitigation than that assessed in
Chapter 4. In this case, some activities, GHGs, and
locations might be subject to mitigation activities
and incentives, while other activities, GHGs, and
locations might not be covered. Many potential
selected activity combinations or mitigation
quantities are feasible. A few are reviewed here
to explore the implications of limiting activities or
quantities of GHG reductions or sequestration:
• setting a fixed national GHG mitigation quantity
for a selected date (e.g., 375 Tg CO2 Eq. per year
in 2025),
• paying for GHG mitigation only for selected
gases (e.g., CO2 only), and
• paying for GHG mitigation only for selected
activities (e.g., agricultural soil carbon only).
This chapter continues first with an analysis of
several hypothetical aggregate national GHG
mitigation levels for the combined forest and
agriculture sectors. The fixed quantities assess-
ment is followed by evaluations of GHG payments
that are limited either in terms of the GHGs
covered, the activities covered, or the prices paid.
Such an approach could be similar in many ways
to project-based mitigation, in which initiators
of a GHG mitigation project take actions to reduce
emissions or increase sequestration on site and
quantify and report these net reductions.
Fixed Quantities of National GHG
The three scenarios evaluated in this section are
defined in Table 5-1. Each scenario sets a fixed level
of reduced net emissions by 375 Tg CO2 (just over
100 Tg carbon) per year below the BAU GHG
baseline for the two sectors by the year 2025.
The three scenarios explore the effect of main-
taining, increasing, or dropping an early, initial
mitigation level in the out years. In the first case
(T-375-375), the 2025 mitigation level is kept in
place thereafter through the end of the projection.
In the second scenario (T-375-900), the 2025
quantity is increased from 375 Tg CO2 to 900 Tg
Table 5-1: National GHG Mitigation Quantity
Scenarios for 2025 and 2055
All quantities are measured in Tg CO2 Eq.
per year net emission reductions below
baseline.
Quantities for 2025 and 2055 can be met by achieving
average annual reductions for the representative decade
(2020-2030, and 2050-2060), respectively.
Scenario U.S. Quantity, 2025 U.S. Quantity, 2055
T-375-375
T-375-900
T-375-0
375
375
375
375
900
0
CO2 (250 Tg C) per year by the year 2055, remain-
ing at that level thereafter. Under the third
scenario (T-375-0), once the 2025 mitigation
quantity is achieved, no aggregate quantity is
specified thereafter. To put this in context, 375 Tg
and 900 Tg CO2 Eq., would respectively offset
about 5 and 13 percent of the U.S. GHG emission
totals for 2003 (EPA 2005).
The analysis uses FASOMGHG to find the solution
to the least-cost combination of activities and
locations to achieve given national mitigation
levels for the forest and agriculture sectors.
National-Level Results by Activity and
Time Period
The results of the FASOMGHG simulations for
the three national mitigation quantity scenarios
are summarized in Table 5-2 and Figure 5-1. They
present national mitigation results that are annual-
ized for the entire 100-year projection period by
activity. These results report the national-level
GHG quantities and marginal cost of the activity
mix that the model identifies as likely to be
implemented to achieve the given GHG reduction
quantity, for the target date, at least cost. Some
key results are the following:
• The scenario that fixes the national mitiga-
tion quantity at 375 Tg per year from year
2025 and beyond achieves that quantity with
a broad mix of activities. While agricultural
soil carbon sequestration and forest management
make the largest contribution, as in the lower-
5-2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
Table 5-2: National Mitigation, by Scenario and Activity, for Least-Cost Quantity in 2025 and 2055:
Annualized over 2010-2110
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline.
Scenario: Quantities in 2025—Quantities in
2055 in Tg CO2 per Year Above Baseline
Annualized (201 0-21 10)
Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All Activities
Marginal Cost per t CO2 Eq. Year 2000 $
T-375-375
18
62
88
35
16
21
240
$23.38
T-375-900
23
70
79
38
20
200
429
$26.10
T-375-0
2
9
54
4
5
0
75
$14.76
Figure 5 1: Least Cost Mitigation Quantities by Scenario and Activity in 2025 and 2055
Quantities areTg CO2 Eq. per year net emissions reduction below baseline, annualized over
2010 2110.
500
Afforestation Forest Agricultural Fossil fuel
Management soil carbon mitigation
sequestration from crop
production
Scenario
Agricultural Biofuel offsets All Activities
CH4 and N2O
mitigation
I T-375-375
I T-375-900 D T-375-0
price scenarios in Chapter 4, the other four
major activities also make substantive contribu-
tions, leading to a diverse portfolio of options.
When the quantity is raised from 375 Tg/
year in 2025 to 900 Tg/year in 2055, the role
of biofuels emerges as a dominant strategy.
In this much larger level of activity emphasizing
longer-term mitigation, biofuels account for
almost one-half the annualized total GHG
mitigation.
When the 375 Tg/year mitigation quantity
level is completely relaxed after 2025, the
policy's effectiveness is substantially under-
mined. It produces less than one-third the
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
5-3
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
annualized mitigation of the constant 375 Tg
quantity (T-375-375) and less than 20 percent of
the T-375-900 scenario quantity.
• Agricultural soil carbon sequestration and
forest management are key options in all
three scenarios. Agricultural soil sequestration
is the first or second contributing activity in all
three scenarios, and forest management is
second or third in all three.
• Afforestation makes little contribution to the
mitigation totals under any of the national
mitigation quantity scenarios. Although
afforestation is a key strategy at the middle
to upper prices in the GHG pricing scenarios
of Chapter 4, the other options are more cost-
effective ways to achieve the fairly modest
national mitigation levels assessed here.
Afforestation is an effective strategy for a more
aggressive effort to achieve higher mitigation
totals at higher cost per unit mitigated.
• The marginal cost ranges from about $15
to $26 per tonne CO2 Eq., depending on the
stringency of the mitigation scenario. The
marginal cost per tonne is about the same for
the scenarios where the mitigation goal stays the
same or rises in the second period (to 2055) but
is about half that amount for the scenario that
has no goal after 2025. The marginal cost of an
additional tonne of mitigation measures the net
cost of an additional unit being added to the
GHG mitigation quantity.1 In essence, this
suggests that additional mitigation could be
warranted if the marginal benefits exceed
these levels.
The summary results of Table 5-2 could mask
important variations in sectoral mitigation over
time. These timing patterns are illustrated in
Figure 5-2, which shows cumulative mitigation
totals over time under the three quantity scenarios,
and in Table 5-3, which reports annual totals by
activity for three key years: 2015, 2025, and 2055.
The patterns demonstrate that the establishment
of fixed and finite-lived mitigation levels can
induce undesirable consequences before the
quantity goal takes effect and after the mitigation
quantity is no longer in place, as described below.
Recall that in each case, the annual mitigation
quantity does not come into effect until 2025.
Therefore, all action in the first decade (2010-
2020) is unrestricted. As a result of this delay, two
phenomena are projected. First, emissions of CO2
and non-CO2 gases are not much affected in the
first decade, because there is no incentive to
achieve these reductions until later. Second, the
sequestration activities reflect anticipatory behav-
ior. The net level of annual sequestration in 2015 is
lower under the national quantity scenarios than
under the baseline, as reflected by the negative
values in Figure 5-2 and Table 5-3. In other words,
the 2025 mitigation quantity goal induces carbon
release in the preceding decade.
The early induced carbon releases are especially
pronounced for forest management, where rela-
tively large carbon reductions are projected in
the decade preceding the mitigation quantity
level taking effect in 2025. This pattern implies a
reaction by forest owners to reduce carbon stocks
before the target takes effect through some combi-
nation of higher harvests or reduced management.
This may be a reaction to preempt some of the
opportunity costs placed on harvests when the
fixed levels take effect in 2025. Nonetheless, it
suggests that a national mitigation quantity set to
take effect a decade or more in the future could
produce some short-run unintended negative
consequences if not designed carefully.
The unintended consequences can extend beyond
the time period as well. For the one scenario in
which the national quantity level is not kept in
place after 2025, net sequestration levels drop
below the baseline for each of the forest and
agriculture sequestration options. Without a
continuing mitigation quantity to shoot for, land-
owners have little incentive to keep carbon stocks
above baseline levels.
1 The cost to consumers and producers is measured as the aggregate sum of producer and consumer surplus in the forest and
agriculture sectors. This is commonly referred to as the "social welfare cost" of a market intervention.
5-4
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
Figure 5 2: Scenarios with Objective of Mitigating: (a) 375 Tg CO2 Eq. in 2025 and Maintaining;
(b) 375 in 2025 and 900 Tg CO2 Eq. in 2055; and (c) 375 Tg CO2 Eq. in 2025 without
Maintaining Thereafter
Quantities are Tg CO2 Eq. cumulative net emissions reduction below baseline to 2110.
Quantities are Tg CO2 Eq. cumulative net emissions reduction below baseline to 2110.
Note: Scale varies for each graph, from 4,000 to 70,000 Tg CO2.
35,000
25,000 -
O
O 15,000
O)
5,000 -
-5,000
D Biofuel offsets
D Crop management FF mitigation
• Forest management
D Ag CH4 and N2O mitigation
D Afforestation
• Agricultural soil C sequestration
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
(a) T-375-375
70,000
D Biofuel offsets
D Crop management FF mitigation
• Forest management
D Ag CH4 and N2O mitigation
D Afforestation
• Agricultural soil C sequestration
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
(b) T-375-900
HI
CM
O
O
O)
D Biofuel offsets
D Crop management FF mitigation
• Forest management
• Ag CH4 and N2O mitigation
D Afforestation
• Agricultural soil C sequestration
-500
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
(c) T-375-0
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
5-5
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
Table 5-3: Least-Cost Mitigation Response to Fixed National GHG Mitigation Levels in 2015, 2025,
and 2055
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline.
Scenario: Quantities in 2025—Quantities in
2055 in Tg CO2 per Year Above Baseline
Year
2015 (midpoint of 2010 decade)
Afforestation
Forest management3
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All Activities
2025 (midpoint of 2020 decade)
Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All Activities
2055 (midpoint of 2050 decade)
Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
All Activities
T-375-375
8
-180
-6
4
5
0
-170
17
234
87
18
19
0
375
3
161
66
59
20
66
375
T-375-900
9
-192
-18
1
3
0
-198
20
230
85
19
20
0
375
33
184
51
69
27
536
900
T-375-0
1
-105
58
3
3
0
-41
9
207
124
17
17
0
375
-13
-22
-99
-3
-2
0
-139
Positive values indicate mitigation or reductions in net emissions below baseline levels. Negative values indicate an increase in
net emissions above baseline levels. Net emission increases are possible when the desired mitigation levels are not in effect,
such as in 2015, and after2025 under!375-0.
5-6 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
Regional Activity Contributions to National
Mitigation Levels
The top 10 region/activity combinations that could
contribute to the national mitigation quantity
scenarios are presented in Table 5-4.2 The region-
activity rankings for the $15/tonne CO2 Eq.
constant price scenario from Chapter 4 are also
listed in Table 5-4 for comparison.
For the two scenarios with mitigation quantity
levels continuing beyond 2025, a diverse mix of
activities and regions comprises the mitigation
portfolio. For the T-375-375 scenario, the top 10
opportunities are spread across eight regions and
across all but one of the activities.
The regional diversity narrows some when the
2055 quantity is set at 900 Tg CO2 per year, because
Table 5-4: GHG Mitigation Quantity Ranking by Region/Activity Combination: Fixed National Mitigation
Quantity Scenarios
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline, annualized 2010-2110.
Scenarios
Region"
CB
SE
LS
SC
NE
SC
RM
SW
NE
GP
RM
NE
SW
CB
SE
SC
RM
SC
SC
Activities
Agricultural soil carbon
sequestration
Forest management
Agricultural soil carbon
sequestration
Fossil fuel mitigation
from crop production
Biofuel offsets
Forest management
Afforestation
Fossil fuel mitigation
from crop production
Forest management
Agricultural soil carbon
sequestration
Agricultural soil carbon
sequestration
Agricultural soil carbon
sequestration
Agricultural soil carbon
sequestration
Afforestation
Biofuel offsets
Biofuel offsets
Agricultural CH4 and
N2O mitigation
Agricultural CH4 and
N2O mitigation
Afforestation
T-375-375
Rank
1
2
3
4
5
6
7
8
9
10
GHG
Quantity
35.6
33.9
31.3
17.4
13.8
12.0
11.8
8.8
7.0
6.8
T-375-900
GHG
Rank Quantity
4 39.3
3 39.9
5 31.6
7 16.9
1 121.7
8 13.5
9 11.8
10 8.8
2 49.3
6 28.8
T-375-0
GHG
Rank Quantity
1 20.8
3 10.4
2 15.2
4 6.8
5 3.8
9 1.6
6 3.6
7 2.0
8 1.9
10 1.5
Constant $15
GHG Price
GHG
Rank Quantity
4 62.2
3 69.2
6 36.9
8 23.7
5 47.9
1 127.7
9 11.7
7 29.3
10 10.5
2 115.8
See Table 3-2 in Chapter 3 for region key.
Consult Chapter 3, Table 3-2, for a key of the regions tracked by the FASOMGHG model.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
5-7
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
5 of the top 10 opportunities occur in the two
southern regions. And agricultural soil carbon
sequestration is the dominant strategy for the
T-375-0 scenario, reflecting the short-term nature
of the scenario. Non-CO2 mitigation is part of the
top 10 set in the T-375-0 scenario only.
National Mitigation Quantity Scenarios
Summary
Taken together, the three national quantity
scenarios provide insights into the importance of
timing in implementing mitigation options. First,
one sees that delaying the achievement of a specific
national mitigation quantity a decade or more can
induce some emitting activity in the short term. This
occurs primarily with the sequestration options,
where carbon stock dynamics inextricably link
actions and carbon consequences across decades.
Therefore, setting a future mitigation goal directly
affects land use and management decisions today.
However, the early reductions in sequestration
found in the model simulations occur, in part,
because these particular scenarios were designed
to achieve annual mitigation quantities, relative to
a future baseline. If, instead, the scenario was set
to maintain a certain level of carbon stock in the
future and this stock was higher than the stock
that exists at the time such a goal is announced,
then the incentive to reduce carbon stocks prior to
the scenario date would be effectively eliminated.
Aiming for cumulative rather than annual mitigation
quantities could potentially avoid these early period
unintended consequences.
GHG benefits are likely to be reversed if the desired
mitigation level is not maintained. But a more perma-
nent enhancement in forest and agricultural
carbon storage and emissions reduction would
require a sustained commitment to achieve
these levels.
Limiting Payments by GHG Type
The analyses to this point have considered all major
GHGs in forestry and agriculture (CO2,CH4, and
N2O) to be subject to mitigation incentives. How-
ever, much of the focus in climate change mitigation
has been on COy whose emissions constitute a
majority of the aggregate anthropogenic global
warming potential, especially in the United States.
Therefore, we consider the consequences of focus-
ing incentives on emissions and sequestration of
CO2 only. This is particularly interesting for the
agriculture sector, a major source of non-CO2
emissions that could be perversely affected by a
CO2-only policy, if it led to increases in agricultural
non-CO2 GHGs.
Paying for CO2 Only vs. Paying for All GHGs:
$15/t CO2 Eq.
To evaluate the CO2-only option, the FASOMGHG
model was run with a price of $15/t CO2 Eq. for
CO2 emissions and sequestration and a price of
zero for the other GHGs tracked by the model.
Results of this scenario are compared to the results
when all GHGs are paid $15 per tonne CO2 Eq. as
illustrated in Table 5-5.
Table 5-5: Mitigation Quantities: Payments for
CO2 Only vs. Payment for All GHGs
($15 pert CO2Eq.)
Quantities are Tg CO2 Eq. per year net
emissions reduction below baseline,
annualized 2010-2110. Net emissions
include non-CO2 gases (even though
payments are for CO2 only).
Activity
Afforestation
Forest management
Agricultural soil carbon
sequestration
Fossil fuel mitigation
from crop production
Agricultural CH4 and
N2O mitigation
Biofuel offsets
All Activities
CO2 Only
110
216
176
49
21
42
613
All GHGs
137
219
168
53
32
57
667
The results in Table 5-5 represent annualized totals
for the entire projection period and can be summa-
rized as follows:
• Limiting payments to CO2 only reduces total
mitigation potential by about 54 Tg/year or
about 8 percent below the mitigation obtained
when all GHGs are priced.
5-8
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
• CO2 and non-CO2 mitigation are largely
complementary:
- About two-thirds of the non-CO2 mitigation
can be accomplished while paying for CO2
only.
- CO2 mitigation (e.g., especially afforestation
and biofuels) is enhanced when non-CO2
gases are included in the payment approach
("all GHGs"), suggesting that non-CO2
reduction incentives divert land from tradi-
tional agriculture to these activities.
- Only agricultural soil carbon sequestration
shows a (slight) trade-off between CO2 and
non-CO2 payments (i.e., the amount of
agricultural soil carbon sequestered declines
very slightly when all GHGs are subject to
payment, rather than just CO2).
The complementarity between CO2 and non-CO2
mitigation is a potentially important factor when
considering incentives for mitigation. First, it
implies that much of the non-CO2 mitigation can
be achieved without explicitly providing incentives
to reduce non-CO2 gases. Second, it implies that
including the non-CO2 reduction activities has
synergistic benefits in CO2 reductions.
CO2 Only: Mitigation Over Time
To illustrate mitigation over time, Table 5-6 pres-
ents the mitigation results for CO2-only payment
by activity for the key years of 2015, 2025, and 2055,
and Figure 5-3 shows cumulative mitigation totals
for the CO2-only and all GHG payment options for
the entire projection period.
The temporal patterns shown in Table 5-6 and
Figure 5-3 reinforce results presented earlier,
namely that forest and agricultural sequestration
options generate sizeable quantities of mitigation in the
first couple of decades after implementation, but these
effects diminish and even reverse in the out years. Also,
as seen previously, the biofuel option does not take
hold for several decades. Figure 5-3 shows that
including non-CO2 GHGs for payment increases
the cumulative mitigation over time but does not
alter the saturation and reversal pattern very
much, because that pattern is driven entirely by
the (CO2) sequestration activity dynamics.
Selected Activity Scenarios
A project-based approach to mitigation is one
in which specific GHG-mitigating activities are
undertaken in distinct locations. One characteris-
tic of project-based approaches is that their scope
is generally limited—some activities are eligible
Table 5-6: National GHG Mitigation Totals in Key Years by Activity: Payment for CO2 Only at $15/t CO2
Eq. (Includes Non-CO2 GHGs)
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline.
Year3
Activity
All Activities
2015
601
2025
627
2055
Afforestation
Forest management
Agricultural soil carbon sequestration
Fossil fuel mitigation from crop production
Agricultural CH4 and N2O mitigation
Biofuel offsets
132
226
201
26
17
0
206
160
209
30
22
0
-180
140
-2
57
17
56
Years represent midpoint of model decades 2010, 2020, and 2050, respectively.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
5-9
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
Figure 5 3: Cumulative Mitigation: Payment for CO2 Only (Includes Non CO2 GHGs) vs. All GHGs
at $15/t CO2 Eq.
Quantities areTg CO2 Eq. cumulative net emissions reduction below baseline.
HI
CM
o
o
01
30,000
2015 2025
2055
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
$15AIIGHG $15 CO. only
for GHG payments and others are not. Therefore, it
is useful to evaluate the effects of a GHG incentive
approach that targets its payments to a selected set
of activities—to see the effect on the sectors'
aggregate mitigation potential, and whether
limiting eligible activities causes unintended
consequences.
GHG mitigation projects can be seen as part of
a broad set of landowner incentive programs
administered by federal or state governments.
There is a long history of these types of programs
at the federal level over the last 50 years. Examples
include the experiences of the Soil Bank Program,
the Forestry Incentives Program (FIP), CRP,
Wetlands Reserve Program (WRP), and various
components of Farm Bill legislation including, of
late, specific provisions to enhance carbon seques-
tration in forestry and agriculture.
To assess such mitigation activity at a smaller scale,
five hypothetical scenarios are defined in Table 5-7.
In each scenario, only one or a small number of
activities receive GHG payments. All other activi-
ties within the forest and agriculture sectors face
no price and thus receive no reward or penalty for
changes in net GHG emissions.
The focus of these scenarios is on activities that
(a) have large potential effects at low prices, as
demonstrated in the results of Chapter 4 (e.g.,
Table 5-7: Selected Activity Scenarios
Activities Subject to Payments
Afforestation
Afforestation + forest management
Biofuels
Agricultural management (agricultural soil carbon +
agricultural CH4 and N2O + crop management fossil
fuels)
Agricultural soil carbon
agricultural soil carbon sequestration, forest
management); (b) are easier to monitor because
they involve a discrete land-use change (afforesta-
tion); or (c) are tied to other closely monitored
market transactions (e.g., biofuels). Although three
of the five scenarios pay for just a single activity,
the other two separately evaluate payments for a
somewhat wider range of forest and agricultural
management activities.
Each scenario is evaluated at one of three price
levels ($/t CO2 Eq.) previously evaluated in
Chapter 4:
• $15, constant over time;
• $3, rising at 1.5 percent per year; and
• $3, rising at 4 percent per year, capped at $30.
5-10
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
National Results
Results of the selected payment simulations are
summarized in Table 5-8. This table shows the
annual mitigation totals in key years (2015, 2025,
2055) for each of the specific activities under each
of the price scenarios. The general patterns across
the activities are similar to those found under
these same price scenarios in Chapter 4 (where
payments were comprehensively applied to all
activities). Agricultural mitigation, specifically
agricultural soil carbon sequestration, is the
primary option at the lowest prices ($3, rising at
1.5 percent), forest carbon sequestration assumes a
large role when prices are somewhat higher ($15,
constant), and biofuels are a key strategy when
GHG prices are expected to rise substantially in
the future ($3, rising at 4 percent per year).
Regional Results
Each of the activities evaluated here has a unique
geographic distribution of mitigation opportuni-
ties in response to the activity-specific GHG
payments. The set of eligible activities, and land-
owner response to GHG price signals, for a given
mitigation incentive is unlikely to be evenly
distributed across regions. The regional implica-
tions and distribution are discussed for each
activity below.
Payments for Afforestation Only
The $15/t CO2 Eq. scenario is the only one of the
three evaluated price scenarios showing much
afforestation occurring at all. Under this scenario,
afforestation is concentrated almost entirely in the
South-Central United States (99 percent of total),
with very small amounts in the Rocky Mountain
and Pacific Northwest regions. Thus, under the
type of targeted afforestation evaluated here,
efforts could be concentrated regionally in the
southern United States, which is where much
of the nation's afforestation and reforestation
are occurring at the present time. Under a more
aggressive policy with higher prices, other regions
would be drawn in as land is competed away from
otherwise more profitable alternatives.
Payments for Afforestation + Forest
Management Only
When forest management is combined with
afforestation for targeted payments, this simulates
the effect of full forest carbon incentives. As shown
in Figure 5-4, this broadening of the incentives
brings in contributions from other regions, for
example, the Pacific Northwest, Westside, and
Northeast. The predominant expansion, however,
is into the Southeast United States, which
Table 5-8: GHG Mitigation under Payment for Specific Activity Scenarios
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline for key years: 2015, 2025,
and 2055.
GHG Price ($/t CO2 Eq.)
$15
Activity Paid for
2015 2025 2055
$3 @ 1.5%
2015 2025 2055
$3 @ 4%
2015 2025 2055
Afforestation
Afforestation + forest management
Biofuels
Agricultural management
Agricultural soil carbon
89 288 -173
350 366 -87
0 0 237
244 242 33
191 184 -39
0
61
0
113
77
0
25
0
129
93
-15a
15
0
51
7
0
69
0
25
-5
0
-58
0
58
16
38
162
352
176
143
Note: Scenarios are not additive because some overlap (e.g., afforestation and forest management).
a Carbon losses from afforestation in 2055 reflect harvesting of forests planted between 2025 and 2055 in this scenario.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
5-11
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
Figure 5 4: GHG Mitigation under Payments for Afforestation and Forest Management Only at $15/t
CO2 Eq.: By Region
Quantities are Ta CO,, Ea. oer vear net emissions reduction below baseline, annualized 2010 2110.
HI
OJ
o
o
01
400
350 -
300 -
250 -
200 -
150 -
100 -
50 -
0
-50
CB LS NE PNWE PNWW PSW RM SC SE
• Afforestation D Forest Management
US
generates about 70 Tg CO2 per year of additional
carbon sequestration through forest management.
The South-Central and Southeast regions together
contribute about 90 percent of the total mitigation
opportunities in the combined forest carbon
scenario, thereby suggesting a fairly concentrated
regional response to forest mitigation opportuni-
ties. This is not surprising given the southern
states' large private timberland base and position
as the nation's largest producer of timber and
forest products.
Payment for Biofuels Only
Consider two points raised in previous chapters
about biofuel adoption: (1) adoption is only
economic at prices of $15/t CO2 Eq. and above and
(2) biofuel demand is assumed to be capacity
constrained in the short run, based on data from
the EIA (Haq 2002). As a consequence, it is not
surprising to find that $3 rising at 4 percent gener-
ates the largest targeted response from biofuel
production of the three price scenarios evaluated.
After about 40 years, the rising price exceeds $15,
and biofuel use capacity is expected to grow
throughout the century (see the biofuel demand
assumptions referenced in Chapter 4). Together
this implies that the capacity expands enough in
time to take advantage of the higher prices. In
contrast, the $15 per tonne constant price attracts
some biofuel adoption over time, but the incentive
does not get stronger as demand constraints relax.
And the $3 per tonne price rising at 1.5 percent per
year is insufficient to draw biofuel production even
in the longer run.
The regional distribution of biofuel production/
mitigation under this price scenario (Figure 5-5) is
a bit wider than the regional distribution of forest
mitigation opportunities, but the concentration is
still entirely within the eastern United States.3 The
Northeast and Corn Belt regions together comprise
about two-thirds of the biofuels opportunity, with
almost all of the remainder in the South-Central
and Southeast regions.
3 Note that Figure 5-5 expresses mitigation quantities as cumulative totals over the entire projection period (2010-2110) rather
than annualized totals. This is done because the discounting and annualization approach presented in Chapter 4 is not appli-
cable under rising-price scenarios (see Herzog et al. 2003).
5-12
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
Figure 5 5: GHG Mitigation under Payments for
Biofuel Offsets Only at $3/t CO2 Eq.,
Rising at 4 Percent per Year, By
Region
Quantities are cumulative mitigation
(2010 2110) in petagrams (billion tonnes)
CO2 Eq.
ONE DCS BSE • SC
D All other
Payments for Agricultural Management
Only
The agricultural management scenario targets
payments for soil carbon sequestration, fossil fuel
(CO2) reductions for crop management practices,
and non-CO2 emission reductions through changes
in crop and livestock management. In Figure 5-6,
the regional distribution of these activities is
depicted under the $15/t CO2 Eq. constant-price
scenario.
The scenario shows that the mitigation activities
are widely distributed across the 10 main agricul-
tural regions in the United States. Much of the
mitigation is the result of agricultural soil carbon
sequestration practices in the Corn Belt, Lake
States, and Great Plains. There is also a modest
amount of mitigation through reductions in fossil
fuel emissions through crop practices in the
South-central and Southwest United States. Non-
CO2 reductions are small, relative to the CO2
options, but comprise a material share of the
Figure 5 6: GHG Mitigation by Region and Activity under Payments for Agricultural Management Only:
$15/t CO2 Eq.
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline, annualized 2010 2110.
<
o
>
o
a.
8
80
70 -
60 -
50 -
40 -
30 -
20 -
10 -
0
-10
« R
n
CB GP LS NE PNWE PSW RM SC SE SW
Ag Soil C Sequestration D Fossil Fuels/Crop Management D Non-CO2 GHG Reduction
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
5-13
-------
CHAPTERS • MITIGATION POTENTIAL OF SELECTED ACTIVITIES
mitigation totals in the Southeast, Southwest,
Rocky Mountains, and Corn Belt.
Payments for Agricultural Soil Carbon
Sequestration Only
The regional distribution of mitigation under the
agricultural soil carbon-only payment scenario for
the $15/t CO2 Eq. constant-price scenario is illus-
trated in Figure 5-7. Landowner responses to the
price incentives are distributed across all agricul-
tural regions, with the Corn Belt generating the
most annual soil carbon sequestration (56 Tg CO2
Eq. per year), followed by the Great Plains (27 Tg)
and Lake States (24 Tg). On the other end of the
spectrum, there is virtually no soil carbon response
(less than 3 Tg CO2 Eq. per year) in the Pacific
Northwest and Pacific Southwest because of
biophysical and economic factors impeding
adoption in those regions at the price trajectory
evaluated. The remaining five regions generate
a modest amount of sequestration in response to
the incentive (between 8 and 11 Tg per year).
Figure 5 7: Regional Distribution of Soil Carbon Sequestration under Payment for Soil Carbon Only:
$15/t CO2 Eq. Constant Price
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline, annualized 2010 2110.
60
I
<5
o.
O
O
oi
50 -
40 -
30 J
20 -
10 -
n
CB
GP
LS
SW
SC
RM
SE
NE
PNWE PSW
5-14
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 6
Implications of Mitigation
via Selected Activities
Chapter 6 Summary
GHG mitigation activities may include project-based approaches (i.e., activity- and location-
specific mitigation actions). Project-based GHG accounting can be used to ensure that the GHG
mitigation attributed to a project reflects its net GHG reductions over time by including baseline
GHG effects that would have occurred without project intervention, reversal of any carbon seques-
tered over time, and any leakage of GHG emissions outside project boundaries. Leakage effects are
found to be more or less confined to the forest sector. The pay-for-afforestation-only scenario
shows leakage of almost 25 percent, whereas leakage appears minimal if all forest carbon manage-
ment activities receive payment. Leakage rates vary regionally and over time because of market
responses and forest carbon dynamics. Most leakage due to targeted afforestation occurs within
the first 2 decades. The broader the spatial scale in which market leakage is evaluated for an activity
that produces commodities traded in that market, the higher the leakage estimated.
Leakage from individual activities in the agriculture sector appears to be small, roughly 0 to
5 percent in this analysis. Paying for additional sequestration through per-tonne CO2 payments is
more efficient than paying on a per-acre basis. Per-acre payments can be made more efficient (i.e.,
more closely match the efficiency of per-tonne CO2 payments) through adjustments based on the
land's carbon productivity potential.
As discussed in Chapter 5, it seems unlikely
for a variety of reasons that fixed limits
would be placed on GHG emissions from
forestry and agriculture. Rather, selected opportu-
nities for mitigation within these sectors may be
seen as an effective means to offset GHG emis-
sions elsewhere. As a result, the scope of eligible
mitigation activities, GHGs, and land coverage
within these sectors may be limited. For the
purposes of this report, these activity- and loca-
tion-specific GHG mitigation actions are called
projects, referring to the actions the landowner
takes on a specific tract of land to mitigate GHGs.
For example, an individual farmer engaged in a
tree-planting activity for the purposes of seques-
tering carbon would constitute a project. This
chapter examines how limiting the scope and
coverage of mitigation actions to project-based
actions can affect the magnitude and distribution
of GHG mitigation within the agricultural and
forest sectors.
Observers have noted a number of important
factors related to implementing these project-
based approaches (CCBA 2004; IPCC 2000):
• demonstrating and quantifying net benefits,
• arranging and paying for the transactions, and
• ensuring sustainable development objectives
are met.
The chapter continues with the discussion of
several key technical issues related to quantifying
GHG benefits, including leakage, baseline-setting,
and permanence or the potential reversibility of GHG
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
6-1
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
benefits. Other project-relevant factors include
measurement, monitoring, and verification (MMV)
of emission reductions or sequestration and
assembly or aggregation of these quantified GHG
benefits across market or program participants.
MMV and assembly can impose transaction costs
that should be considered when evaluating the
economic attractiveness of mitigation projects.
These issues are all discussed in more detail in the
section that follows.
Because it is an aggregate model operating at
regional resolution, FASOMGHG does not directly
model implementation of activity at the individual
project level. However, the model is flexible enough
to limit the scope of incentives to subsets of activi-
ties, regions, and GHGs, thereby providing some
insight into the effect of such limitations on
mitigation potential. For instance, FASOMGHG is
used in this chapter to estimate leakage potential
when GHG incentives are confined to a subset of
activities. In addition, the chapter includes an
empirical analysis of modifying how incentives are
provided to assess GHG payments on a per-acre,
rather than per-tonne (t CO2), basis (the approach
thus far). Per-acre payments have been discussed
as a means to economize on MMV and transaction
costs (Antle et al. 2003).
The next section further discusses project-level
implementation issues and the extent to which
these factors can affect a project's net GHG
benefits.
Project Quantification Issues and Costs
Project-based GHG mitigation activities are
typically defined as those with clearly defined
geographic boundaries, time frames, and institu-
tional frameworks (IPCC 2000, Chapter 5). Certain
characteristics of forestry and agricultural project
activities can complicate the estimation of their net
GHG mitigation benefits. Methods to address
these concerns are discussed below.
Quantifying the Net GHG Contribution
of Projects
One challenge with project-based approaches is
ensuring that the amount of mitigation attributed
to a particular project reflects the net contribution
of that project to GHG reductions over time. Of
particular importance is the notion that the GHG
accounting captures
• the (baseline) GHG effects that would have
occurred without the project intervention;
• the reversal or re-release of any carbon seques-
tered over time through harvesting, discontinu-
ation of practices, or natural disturbance; and
• any leakage of GHG emissions that may have
occurred outside the boundaries of the project.
Each of these issues is addressed below. Special
emphasis is placed on the leakage issue because
FASOMGHG model simulations in this report, in
addition to other recent studies, are able to quan-
tify leakage effects from activity-specific incentive
programs.
Establishing Project Baselines
The net GHG benefit of mitigation at the project
scale can be estimated as the additional GHG
emission reductions (sequestration) that occur
relative to emissions (sequestration) levels in the
project's absence. This is the concept of additional-
ity. To determine additionality, one can estimate
what would happen under business-as-usual or
BAU without the project, which is referred to as the
project baseline (IPCC [2000], Chapter 5).
A number of analyses and existing GHG mitiga-
tion programs have focused on the primacy and
complexity of setting a baseline case to estimate
GHG mitigation benefits (e.g., IPCC [2000],
Chapter 5). Demonstrating additionality requires
establishing a project baseline. In the case of GHG
emission reduction projects in sectors such as
electricity generation, a baseline might reflect the
GHG emission rate that would prevail if the
electricity were generated using standard
technologies and fuels for a given sector and
region. In forest- or agricultural-sector projects,
however, it is a bit more complicated. First, an
estimation of the land-use practices that would
occur under BAU may be required. This may
require using historical data on land use and
management practices to provide an empirical
6-2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
foundation for BAU. The emergence of remotely
sensed land-use data in a digitized format expands
the possibilities for more complex and rigorous
analysis of baseline land-use behavior. Then, once
the land-use or management practice baseline is
determined, estimation of what the emissions or
sequestration rate would be under each of the BAU
land-use practices can complete the baseline
quantification.
No generally agreed methodology yet exists in the
United States or internationally for project base-
line setting by activity and region, although
numerous efforts are under way to develop consis-
tent protocols (CCBA 2004). It is beyond the scope
of this report to assess project-level baseline
options. Those methods are still largely in the
proposal and evaluation stages. However, the
development of project baselines is a cost of
project development that is not directly captured
in the economic analysis herein. This and other
potential project transaction costs are addressed
further below.
The focus of the discussion in this section has been
on baselines at the project level, but sector-level
baselines also are used in the broader analyses
presented in Chapters 4 and 5. All mitigation
results in the report are presented relative to
the FASOMGHG sector baselines for forestry
and agriculture. Thus, they are consistent with the
concept of additionality discussed here. However,
the model scenarios in those chapters do not
impose additionality as a requirement for GHG
payment—in essence all GHG effects are
potentially eligible for payment.
Duration and Potential Reversal of GHG
Benefits (Permanence)
As discussed throughout this report, GHG miti-
gation in the forest and agriculture sectors is
susceptible to reversal. This is particularly relevant
when carbon is sequestered for some time and
then re-released accidentally (e.g., through wild-
fire) or as part of a planned intervention such as
harvesting or land-use change. A complete ac-
counting framework would capture both GHG
releases to and GHG removals (sequestration)
from the atmosphere. The FASOMGHG model
scenarios presented in this report do capture such
carbon losses from intentional releases tied to the
harvesting and land-use decisions embedded in
the model. Accidental carbon releases through fire,
insects, and diseases are captured in the model via
the biophysical yield functions used for forestry
and agriculture, which are generally based on
average yields, and therefore implicitly capture the
persistent accidental losses from ambient sources.
However, a number of logistical factors may make
such a complete accounting of GHG releases and
removals over time as modeled in FASOMGHG
for this report difficult for individual forestry and
agriculture projects. These factors revolve around
two key questions: (1) how does a set of mitigation
activities or individual projects address the risk of
reversal of GHG benefits during the lifetime of the
program, and (2) how does it address this risk of
reversal once the program or project has ended?
Specific factors to consider include the following:
• Natural disturbance and other force majeure
effects occur with uncertainty.
• Catastrophic loss of carbon could cause
catastrophic financial losses for an investor.
• Project contracts generally have finite lives.
The first two factors relate to the difficulty of
dealing with the risks of release when the project
is under way. The unpredictability of project risk
complicates project planning and decisions on
actions that might be taken to reduce risks. By and
large, the prospect that the investor might suffer
catastrophic loss of the asset—carbon benefits,
plus the normal accompanying economic asset,
such as timber—makes the investment more risky
and therefore reduces its attractiveness. If the risks
are large enough, investors may seek ways to cover
these potential losses if they proceed with the
investment. Specific instruments for covering
these risks (insurance policies, pooling projects
with similar or dissimilar characteristics, holding
some achieved mitigation benefits in reserve)
might be considered, although the markets for
these financial instruments may be a bit thin at
this time (Subak 2003).
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
6-3
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
The other critical issue is that the project will
typically involve a contract that expires after some
period of time. The question then arises: how do
you account for risks of release after a project
ends? Various parties have proposed contractual
options to address the risk of reversal in (primar-
ily) carbon sequestration projects. These options
are described in Table 6-1.
The options in Table 6-1 address how to account for
reversal when it occurs. But project developers may
also want to consider the actions they can take to
minimize the risk of GHG reversal at the project
design stage. One approach is to develop a carbon
reversibility management plan, which lays out
steps for identifying reversal risks, evaluating
options for minimizing these risks, developing
liability or compensation for risk when it occurs,
and monitoring risks over the life of the project
(WRI-WBCSD 2003).
Analytic consideration of project reversibility is
outside the scope of this analysis and remains a
topic of continued dialog and research.
Assessing the Potential for Leakage
Project-based mitigation approaches run the risk
that some of the direct GHG benefits of these
efforts will be undercut by leakage of emissions
outside the boundaries of the project. IPCC (2000)
defines leakage as "the unanticipated decrease or
increase in GHG benefits outside of the project's
accounting boundary (the boundary defined for
the purpose of estimating the project's net GHG
impact) as a result of project activities." The notion
that project-based mitigation can generate leakage
is a widely accepted concept.
Table 6-1: Candidate Approaches for Accounting for Reversal Risk from Carbon-Based GHG Mitigation
Projects
Comprehensive accounting
Pay-as-you-go
Used in this report
with FASOMGHG
model
Accounts for both carbon storage
and carbon release to the atmosphere.
This approach is consistent with national
GHG inventory accounting practices.
Addresses reversal as long as activity is
reported in continuous program, including
reversal beyond the finite life of a project.
IPCC (1996, 2000); Feng et al.
(2001)
Approaches to project reversal risk (if comprehensive accounting not used)
Temporary crediting
Designed to account explicitly for the fact
that sequestration projects may only yield
temporary reductions in atmospheric CO2
concentrations.
Three general approaches:
• expiring, or temporary, Certified
Emission Reductions, or tCER;
• carbon "rental"; and
• carbon "leasing."
Colombian Ministry of the
Environment (2000); Blanco
and Forner (2000); Chomitz
(2000); Marland et al. (2001);
Moura Costa (1996); Dutschke
(2001 );Dutschke (2002)
Ex ante discounting
Directly estimate and account for
predicted reversal through
management, harvesting, etc., in
determining sequestration tonnes
assigned at the beginning of the project.
McCarl and Murray (2002);
Lewandrowski et al. (2004)
6-4
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
the system. Leakage is calculated as a percentage
of the direct benefits, accordingly:
Indirect GHG emissions
Leakage from nontargeted activity
percent Direct GHG reductions
from targeted activity
xlOO.
The challenge is quantifying leakage attributable
to a specific activity and location. Leakage is
relevant for assessing the effectiveness of pro-
grams that target a subset of land-based activities
such as afforestation, biofuels, or agricultural soil
carbon sequestration, as in the case of the scenar-
ios presented in Chapter 5. Therefore, it is impor-
tant to recognize the potential for leakage and to
develop methods to
• target or design projects or sets of mitigation
activities to minimize leakage,
• monitor leakage after projects or sets of mitiga-
tion activities are implemented,
• quantify the magnitude of leakage when it
exists, and
• take leakage into consideration when estimating
net GHG benefits of activities.
There has been little quantification of leakage
effects in the forest and agriculture sectors.
Chomitz (2002) uses an analytical model to com-
pare the potential for leakage from forestry proj-
ects to that from energy-sector projects. Chomitz
shows that forestry projects are not systematically
more prone to leakage than energy-sector ones, as
some parties have argued.
The five selected activity scenarios presented in
Chapter 5 provide a framework by which to
estimate the extent of leakage from selected, non-
comprehensive activity sets. In each case, only one
activity or subset of activities receives GHG
payments. The GHG mitigation from each activity
is then quantified and presented as the direct
benefits of a selected activity. Although payments
may only be applied to a single activity or subset,
the FASOMGHG model tracks GHG effects
throughout the entire U.S. forest and agriculture
sectors. Therefore, one can compare the direct
GHG benefits of each set of targeted payments
with the net GHG effects for the entire combined
sectors to quantify if and to what extent the direct
benefits are offset by leakage somewhere else in
1 Leakage effects in Table 6-2 are presented for the $15/tonne CO2 Eq. price because that price induces some activity in all
categories. The lower prices evaluated in Chapter 5 ($3/tonne, rising at 1.5 percent and 4 percent per year) generate too little
afforestation to discuss leakage effects for that activity.
As has been demonstrated throughout this report,
GHG mitigation actions in forestry and agriculture
generate variable levels of mitigation over time,
particularly for the sequestration options. To
capture these fluctuating GHG effects in a single
measure of leakage for each activity, the GHG
quantity terms in the numerator and denominator
of the leakage equation are expressed in annual-
ized equivalent values for the corresponding
projection period, decades 2010 to 2110. The
implications of choosing a shorter time horizon for
leakage estimation are discussed further below.
Table 6-2 presents the corresponding leakage
estimates for each of the selected activity scenar-
ios, evaluated at a single GHG price of $15 It CO2
Eq.1 for each of the FASOMGHG-selected activity
scenarios from Chapter 5. The most significant
finding is that only one of the activities, affores-
tation, generates appreciable amounts of leakage
(24 percent).
Once afforestation and forest management are
combined and targeted together, almost all of the
leakage vanishes because essentially all of the
leakage from mitigation incentives that induce
afforestation occurs through carbon reductions
from reduced forest management. This reduced
forest management is caused by the corresponding
decline in timber prices and incentive to invest in
forest management caused by increasing the area
of land in forests. When forest management is
eligible to receive incentive payments, this leakage
largely goes away. In fact, the leakage effect is even
slightly negative, meaning that there is a small
amount of "good" leakage (reduced net emissions)
spilling out of the forest sector into the agriculture
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
6-5
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
sector, further augmenting the benefits of the
direct payments for forest carbon. This good
leakage occurs as the sectors reallocate land and
management in response to the forest-sector
incentives, and the reallocation of resources in
agriculture leads to a slight decline in agricultural
emissions (i.e., an increase in indirect mitigation).
These leakage values are small in both absolute
and percentage terms. Given the uncertainty
involved in any complex modeling exercise as this,
the more important message is that leakage
appears minimal if all forest carbon activities are
targeted for payment together. Likewise, the
results in Table 6-2 suggest that leakage from
payments targeting biofuels and agricultural
activities is quite small, as well, roughly 0 to
6 percent.
The time horizon for GHG mitigation, particularly
forest carbon sequestration, is long, with actions
taken in one year having implications for many
decades down the road. However, the time horizon
for projects or sets of reported mitigation activities
is likely to be shorter, confined by the institutional
realities of changing policy priorities and of
investment time frames. The discussion in Box 6-1
considers the implications of viewing leakage
effects for an afforestation project from a shorter
time frame than the 100-year projection period
used to generate the leakage estimates in Table 6-2.
It concludes that for the afforestation $15/t CO2
scenario reviewed, the leakage rate is unchanged
from the 100-year value under a 50-year time
frame of analysis. But it significantly increases
under a 20-year time frame because most affores-
tation leakage occurs in the first few decades.
Leakage from Forest Carbon Sequestration:
A Closer Examination
Because the results in Table 6-2 suggest leakage
effects are more or less confined to the forest
sector, we take a closer look at forest carbon
leakage, further detailing the FASOMGHG results
and drawing from other published forest carbon
leakage estimates.
Focusing first on the leakage results from paying
for afforestation only, the 137 Tg CO2 per year of
direct GHG benefits from afforestation is offset by
leakage of about 33 Tg CO2, or about 24 percent.
Thus, the net GHG benefit is 104 Tg CO2, when
leakage is taken into account.
Table 6-2: Leakage Estimates by Mitigation Activity at a GHG Price of $15/t CO2 Eq.
All quantities are on an annualized basis for the time period 2010-2110.
Selected
Mitigation Activities
A
GHG Effects
of Targeted
Payment
(Tg C02 Eq.)
B
Net GHG
Effects of
All Activities
(Tg C02 Eq.)
Indirect GHG
Effects from
Nontargeted
Activity"
(Tg C02 Eq.)
Afforestation only
Afforestation + forest management
Biofuels
Agricultural management
Agricultural soil carbon
137
338
84
230
154
104
348
83
231
145
-33
10
-1
1
-9
24.0
-2.8
0.2
-0.1
5.7
a Indirect effects: C = (B - A).
b Leakage rate: D = -(C/A) x 100; rounding occurs in table.
Note: Negative leakage rate in D refers to beneficial leakage (i.e., additional mitigation outside the selected activity region,
also called positive leakage).
6-6
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
In what activities and regions can the leakage be
found? Figure 6-1 provides some insights. As
described in Chapter 5, virtually all of the affores-
tation response in the afforestation-only payment
scenario occurs in the South-Central states (about
99 percent). This is depicted in the left side of
Figure 6-1. The right side of Figure 6-1 shows the
regional and activity nature of the leakage induced
by the afforestation payments. The primary source
of leakage is, as expected, from the decline in
carbon from forest management. But Figure 6-1
shows two other nonforest leakage effects caused
by the movement of land from agriculture to forests
within the South-Central region. First, this land
movement produces a decline in crop-related fossil
fuel (CO2) emissions within the region, which is
Box 6-1: Shortening the Time Horizon for Quantifying Leakage
The leakage estimates in Table 6-2 are calculated using
the annualized values for the time stream of GHG
mitigation effects over the entire FASOMGHG projection
period, spanning the time period 2010 to 2110. These
annualized values capture in one summary metric the
entire projected mitigation profile over a long period of
time. However, analysts also might be interested in
confining measurement of leakage just to a set period of
time pertinent to a given mitigation reporting framework
(e.g., 2010) or the time frame of a given project. This
may be particularly applicable to highly time-dynamic
mitigation options such as afforestation. Therefore, we
recalculate the leakage estimates for the afforestation
scenario, confining the time period of observation to 5
decades and 2 decades, respectively, and ignoring all
future GHG effects beyond that. The effect of the
change in time horizon is reflected below for the $157
tonne CO2 Eq. GHG price.
Targeted Mitigation Activity: Afforestation at $15/t CO2 Eq.
All GHG quantities in the table are annualized over the time horizon indicated in far left column.
Indirect GHG
Effects from
Nontargeted
Activity"
(Tg C02 Eq.)
Leakage Time
Horizon
GHG Effects
of Targeted
Payment
(Tg C02 Eq.)
B
Net GHG
Effects of
All Activities
(Tg C02 Eq.)
D
Leakage
Rate"
10 decades
5 decades
2 decades
137.4
170.7
208.5
104.4
129.7
127.7
-33.0
-41.0
-80.8
24.0%
24.0%
38.8%
a Indirect effects: C = (B - A).
b Leakage rate: D = -(C/A) x 100; rounding occurs in table.
Note: Negative indirect effects produce positive leakage rate.
Shortening the time horizon from 10 to 5 decades,
while it affects the absolute annualized GHG mitigation
quantities, does not affect the relative leakage rate. In
essence, most of the important feedbacks between
afforestation, forest management, and other activities
are resolved in the first 5 decades.
However, when the time horizon is shortened to just
2 decades, both the absolute annualized mitigation
values and the leakage rate are substantially affected.
The leakage rate goes up because the initial response
to an afforestation incentive payment is a decline in the
area and intensity of managed forests not subject to the
afforestation payments. This decline leads to a large
drop in carbon on these other managed forests in the
initial decades, which eventually evens out.
However, when the time horizon is confined to
2 decades, these initial declines in forest management
carbon have a larger effect relative to the direct affores-
tation GHG benefits, which will continue to accumulate
for several more years after the second decade.
This exercise suggests that most of the leakage effect
from an afforestation project occurs in the first couple of
decades. Therefore, if any project-level accounting
standard chooses to ignore all carbon effects beyond
the second decade, leakage effects will appear to be
higher than their projected effect over a longer time-
frame.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
6-7
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
shown in Figure 6-1 as positive mitigation (i.e.,
"good" leakage). Second, this land movement
reduces the South-Central cropland base and leads
to more intensive cultivation practices, which
increase soil carbon loss in the region (i.e., "bad"
leakage).
The phenomena depicted in Figure 6-1 imply that
this afforestation scenario, which turns out to be
regionally confined to the South-Central United
States for the scenario evaluated, has leakage
effects that are also regionally confined. Virtually
all of the leakage occurs within the two southern
(South-Central and Southeast) regions. Most of
the market feedback from this level of afforesta-
tion would have spatial limitations, because land-
use change has localized tendencies. Forest
management responses are confined to the South-
Central and Southeast regions, because that is
where most of the country's intensively managed
forests are located.
Leakage Estimates from the Literature
A study by Murray, McCarl, and Lee (2004) uses
FASOMGHG's precursor, the FASOM model, to
estimate leakage from different U.S. forest carbon
sequestration activities. Other than using the same
basic modeling foundation, the Murray et al. study
differs from this report in a number of ways. For
example, the Murray et al. study includes scenari-
os for forest preservation and avoided deforesta-
tion in addition to afforestation but does not
estimate leakage from agriculture or biofuel
production.2 That study also tries to simulate
Figure 6 1: Regional Leakage Flows for Afforestation Only Payment Scenario: $15/t CO2 Eq.
Afforestation Induced
by Payments
Leakage Effects
140
SC All
Afforestation Other
Afforestation
SC Fossil SC SE Forest SC Ag. All Other
Fuels/Crops Forest Mgmt. Soil C Activities
Mgmt. Sequestration and
Regions
Note: Negative sign (e.g., South-Central Forest Mgmt.) is leakage, and positive sign is beneficial leakage (i.e., additional mitigation
outside targeted activity region).
2 Forest preservation refers to the withdrawal of existing forest from the timber harvesting base, also referred to sometimes as a
forest set-aside. Avoided deforestation refers to keeping land in forest that would otherwise be converted to another use. Once
deforestation is avoided, the forest can either be preserved (no timber harvesting allowed) or maintained as a timber-producing
forest with harvests allowed.
6-8
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
smaller, region-specific mitigation incentives, in
contrast to the national-level payment scenarios
evaluated here. Their leakage estimates are
derived by simulating a specific level of mitigation
in a given region for a single activity and then
comparing model results for that selected activity
level to the United States as a whole. They assess
forest set-asides or preservation of lands likely to
remain in forest (100,000 acres of old growth in the
PNW and 600,000 acres in the South), avoided
deforestation on lands with potential for conver-
sion to agriculture, and afforestation (a 10-million-
acre level in each region). These two studies taken
together can provide some sense of the range of
forest carbon leakage estimates in the United
States by activity and region.
The national afforestation estimate in Table 6-2
(24 percent) falls in the 18 to 43 percent range
found for regional leakage in U.S. afforestation by
Murray et al. (see Table 6-3). But in contrast to this
study, where afforestation generates the largest
leakage of any of the activity scenarios evaluated,
Murray et al. find in some cases larger leakage esti-
mates for the other forest-sector activities: forest
preservation and avoided deforestation (Table 6-4).
Table 6-3: Afforestation Regional Leakage
Estimates from Murray et al. (2004)
Region
Northeast
Lake States
Corn Belt
Southeast
South-Central
Leakage %
23.3
18.3
30.2
40.6
42.5
Forest preservation leakage was found to vary from
16 percent in one region (PNWW) to almost 70
percent in another (South-Central). Forest preser-
vation can generate relatively high leakage if it
simply shifts harvests to another location, which
is what the results for the South-Central region
suggest. There is less leakage from preservation
in the PNWW, in part, because the harvests are
shifted to other regions where the losses in carbon
would not be as high as they are in the carbon-rich
forests of the Pacific Northwest.
Leakage for avoided deforestation is found to
vary from slightly positive leakage (i.e., net positive
GHG effects off-site) in the Corn Belt, to about 8
Table 6-4: Forest Preservation and Avoided Deforestation Regional Leakage Results from Murray et al.
(2004)
Forest Preservation (Set-aside)
Pacific Northwest-Westside (PNWW)
South-Central (SC)
Leakage %
16.2
68.8
Leakage %
Harvesting Allowed on Preserved Forests?
Region
Avoided Deforestation
Pacific NW-East Side (PNWE)
Northeast (NE)
Lake States (LS)
Corn Belt (CB)
South-Central (SC)
No
8.9
43.1
92.2
31.5
28.8
Yes
7.9
41.4
73.4
-4.4
21.3
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
6-9
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
percent for the PNWE, to leakage topping
40 percent in the Northeast and Lake States, where
it reaches 73 to 92 percent. Leakage is higher when
no harvesting is allowed on the lands saved from
deforestation, as harvests are shifted to other
forests as described above.
Other studies in the literature do not address GHG
leakage directly but focus on the market activity-
shifting that underlies GHG leakage. For instance,
Wear and Murray (2004) used an econometric
model of the U.S. softwood lumber market to
simulate the effect of reducing timber sales in the
Pacific Northwest. Federal restrictions on the
harvest of old-growth timber in the 1990s resulted
in an 85 percent reduction in harvest volume on
public lands. Wear and Murray found that 43
percent of timber harvest reductions in the West
region alone leaked away into other harvests
within the region, that 58 percent leakage occurred
when the continental United States was consid-
ered, and that fully 84 percent of the leakage
occurred when the United States and Canada were
included in the analysis.
In the area of agricultural soil management,
previous work by Wu (2000) and Wu, Zilberman,
and Babcock (2001) examines program "slippage"
from CRP adoption in the United States. Slippage
refers to the phenomena by which land retirement
into the CRP can induce lands outside the program
to enter into cultivation and offset the direct
benefits of land retirement. These studies find that
10 to 20 percent of direct CRP benefits are offset by
slippage. The agricultural soil carbon sequestra-
tion leakage estimate in this study (5.7 percent) is
slightly below, but in the same ballpark as, those
slippage estimates.
Leakage Summary
Several key findings emerge on leakage from both
this study and the extant literature.
First, afforestation, forest preservation, and
avoided deforestation, if targeted individually,
could have significant to very large leakage—
depending on the region and how incentives for
mitigation are provided. The forest economy
involves multiple feedbacks between markets for
land, other inputs, and timber. So when GHG
incentives are confined to just one part of the forest
production system—land use, management,
harvest timing—it is more than likely that another
part of the system will be affected, often in ways
that diminish the net GHG mitigation for the
entire forest system. For instance, when afforesta-
tion is awarded GHG price incentives and forest
management is not, then forest management
intensity and carbon tend to decline. Likewise,
when harvests are restricted in certain areas but
allowed to vary freely elsewhere, the market will
tend to shift the harvests and cause leakage.
Second, this key finding follows directly from the
first, namely, leakage appears minimal if all forest
carbon activities are included for payment together.
For instance, if afforestation and forest manage-
ment are targeted together, very little leakage
occurs because leakage from afforestation occurs
through carbon reductions from reduced forest
management. Forest management is reduced
because of the corresponding decline in timber
prices and incentive to invest in forest manage-
ment. When incentives are provided to forest
management, "good" leakage may occur as the
sectors reallocate land and management in
response to the forest-sector incentives, and the
reallocation of resources in agriculture leads to
a slight decline in agricultural emissions.
Third, leakage from individual activities outside
the forest sector appears to be small. The results
in this study suggest that leakage from payments
targeting biofuels and agricultural activities is
quite small, roughly 0 to 5 percent. Therefore,
any accounting adjustments for leakage could
fall more heavily on forest-sector activities than
on agriculture.
Fourth, leakage varies by region for a given mitiga-
tion activity, reflecting differing levels of market
response for wood products or other commodities
within and across regions.
Fifth, leakage rates vary over time because afforest
carbon dynamics; therefore, leakage estimates may
6-10
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
vary depending on the time frame of analysis.
FASOMGHG results here show that most leakage
due to targeted afforestation occurs within the first
2 decades.
Finally, while only early analyses are available to
date, it appears that the broader the spatial scale in
which market leakage is evaluated for an activity
that produces commodities traded in that market,
the higher the leakage estimated. The FASOMGHG
model does not capture leakage due to GHG
incentive responses outside the United States.
However, the FASOMGHG results in this study
show that, at least for afforestation, leakage may be
relatively confined to within the regions directly
affected by incentives for mitigation. For harvest
restrictions, the spatial scale is wider, because the
results of Wear and Murray (2004) clearly show
higher leakage rates as the number of regions in
the North American timber market included in the
analysis increased. Therefore, a more global view
is needed to better assess mitigation activities and
incentive approaches that might cause shifts in
production to other regions of the world.
Other Project Implementation
Considerations
A number of other implementation issues should
be considered when evaluating project-based or
other selected activity approaches to GHG mitiga-
tion in the forest and agriculture sectors. These
implementation issues are reviewed below and are
not explicitly reflected in the FASOMGHG scena-
rios throughout this report.
Measurement, Monitoring, and Verification
(MMV)
MMV is the process by which the amount of GHG
mitigated by a project is measured, the measure-
ments are monitored over time to ensure that all
relevant GHG flows are accounted for, and the
monitored measurements are verified to demon-
strate to external parties that the emission reduc-
tions and/or sequestration have occurred. For
carbon sequestration projects, this process can
involve a range of methods, including repeated
measurement of sample plots using refined scien-
tific procedures, collection and analysis of aerial
photographic and satellite image data, and use of
ecosystem process models to simulate likely
outcomes when observation is difficult.
The ability to measure GHG effects in forestry
and agriculture depends a great deal on the
• GHG of interest,
• number and location of affected carbon storage
pools,
• way in which the GHGs are exchanged between
ecosystems and the atmosphere,
• precision that is acceptable for reporting and
verification purposes, and
• cost one is willing to pay to develop the
measurements.
For instance, the amount of carbon stored above
ground in trees is relatively easy to measure,
but the amount of carbon stored in soils is more
difficult. Detecting the change in soil carbon can
generally be more difficult because of a high
degree of spatial variability and the fact that any
change may be small relative to the size of the
existing soil carbon stock. See the following for
more detail on MMV issues for forestry and
agricultural sequestration projects: Chapter 5
(e.g., Table 5-7) in IPCC (2000); CASMGS (2003)
Carbon Measurement and Monitoring Forum at
www.oznet.ksu.edu/ctec/ Fall_Forum.htm; and
Brown (2002).
CH4 emissions from livestock enteric fermentation
are difficult to measure at the herd level, but
monitoring CH4 emissions avoided through
manure management systems that use the CH4
for energy production is relatively easy, because
the CH4 is directly tied to the amount of kWh
produced. Likewise, CO2 emissions reduction
from replacing fossil fuels with biofuels is a
relatively straightforward measurement because
of its correspondence to actual, observable market
transactions. In light of these factors, MMV
requirements need to be taken into consideration
before embarking on a project, because this can
affect the ability to demonstrate credible mitiga-
tion effects and can substantially affect the cost
of the project.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
6-11
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
Market Assembly and Brokering of
Mitigation Activities
For a GHG mitigation market to work, buyers and
sellers must be brought together to consummate
transactions. Some process is necessary by which
GHG mitigation benefits are assembled and
brokered. Without this, the economic incentives
for mitigation may not flow to those who can
supply the mitigation at a cost that is less than
or equal to the price that a buyer is willing to
pay. When there are few numbers of buyers and
sellers (i.e., the market is thin), this may create
an inefficient process of search and discovery.
When there are more market participants, a role
for third parties to broker and assemble transac-
tions could evolve. Consequently, the development
of this market-making infrastructure may need
to be considered in any market-based GHG
mitigation program.
Even in the case of government-sponsored land-
owner incentive programs, rather than a private
market for mitigation, some infrastructure is
necessary for delivering the incentive to the
landowner. In the United States, there is a long
history of these programs being delivered to
farmers, ranchers, and forestland owners through
a variety of outreach mechanisms such as agricul-
tural and forestry extension programs at federal
and state agencies and universities.
Transaction Costs
The various implementation issues just discussed
(e.g., contracting, risk management procedures,
MMV, market assembly) all impose what can be
termed collectively as transaction costs on devel-
oping and operating a GHG mitigation project.
The liability for these transaction costs may fall on
the buyer, the seller, or both parties.
If the seller is liable, this adds to their costs and
increases the amount they need to be compensated
to voluntarily engage in the transaction. If the
buyer is liable, this lowers the amount they are
willing to pay for a unit of mitigation, because the
full cost of the unit includes the transaction cost.
But regardless of who bears the direct liability, the
cost and risk of undertaking these activities
directly affect the value of the transaction
itself.
Many of these transaction costs operate under scale
economies; that is, because they involve many
costs that are largely fixed, the cost per transaction
declines with the number of transactions covered
(Mooney et al. 2004b). For example, a reversal risk
management plan and MMV plan will not likely be
10 times larger for a project generating 100,000 t
CO2 Eq. per year of mitigation than one that gener-
ates 10,000 t CO2 Eq. per year. In addition, GHG
contracts may need to be bundled or aggregated
to a minimum lot size for market exchange. For
instance, the Chicago Climate Exchange, a volun-
tary system for GHG trading, requires a minimum
trading block of 12,500 t CO2 Eq. If conservation
tillage practices generate 0.5 t CO2 Eq., per acre per
year, this will require bundling across 25,000 acres.
Therefore, large operations will be able to bundle
more cost-effectively than small ones. Finally,
market assembly or brokering costs are likely to be
much lower on a per-unit basis for a large volume
market than for a small volume market. Note that
the absolute size of the transaction costs per unit
does not matter as much as the ratio of that cost
to the per-unit value of the transaction.
Evidence on the size of transaction costs associated
with forest and agricultural practices is quite
limited. Relatively few GHG mitigation projects
in forestry and very few in agriculture have been
implemented in the field. Certain components,
such as the cost of MMV, have been recorded in
some cases and have been relatively low for proj-
ects operating on a fairly large scale. Kadyszewski
(2001) estimates costs of less than $0.25/t C Eq.
($0.07/t CO2 Eq.) for forest carbon measurement.
Mooney et al. (2004a) estimate the measurement
and monitoring costs of soil carbon benefits from
the adoption of more intensive cropping practices
in Montana as generally less than $l/t C Eq. ($0.30
per t CO2 Eq.). However, costs will depend primar-
ily on the degree of precision required, heteroge-
neity of the landscape, frequency of sampling, and
project size (Mooney et al. 2004a; Brown, Masera,
and Sathaye 2000).
6-12
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
While measurement costs may be low, on the other
hand anecdotal evidence suggests that some
transaction cost components could be considerable.
For instance, if trading tends to be conducted in
large units (e.g., 100,000 t CO2 blocks), given the
sequestration rates per unit of output for many
of the activities in forestry and agriculture, each
transaction could require aggregating hundreds or
thousands of landowners. These costs are likely to
be considerable. Alston and Hurd (1990) found that
the costs of delivering government programs to
farmers in the United States are on the order of 25
to 50 percent of the value of the program payments.
The FASOMGHG model simulations throughout
this report do not include transaction costs. This
is not problematic if transaction costs are low,
because their omission from the analysis would
then be trivial. If transaction costs are uniform
across options, then one can adjust the GHG price
incentives accordingly and roughly determine the
mitigation potential. On the other hand, if per-
unit transaction costs differ among afforestation,
forest management, agricultural soil carbon
sequestration, and biofuels, then the portfolio of
options selected at each GHG price will change.
Consistent data on the size and distribution of
transaction costs across mitigation options would
be a helpful addition to analyses such as those
presented in this report.
Preliminary Assessment of Implementation
Factors by Major Mitigation Activity
The discussion above suggests that major mitiga-
tion activities have different characteristics with
regard to project-based implementation. Tables 6-5
and 6-6 evaluate mitigation options across the
various implementation issues, quantitatively
where FASOMGHG results are available, and
qualitatively otherwise. A rigorous comparison
of activities along each of the implementation
factors requires additional analysis and is beyond
the scope of this study. A review of Tables 6-5 and
6-6 suggests the following:
• Afforestation has significant leakage varying by
regional market conditions, but MMV and
establishment of a baseline may be relatively
straightforward because land-use change can
be observed. Additionality is likely to be high.
Reversal risk is relatively high without
constraints imposed.
• Forest management, which is an economic
option at a wide range of options, has some
project implementation challenges. MMV and
baseline setting may be more challenging than
afforestation, for example, because changes
in management practices rather than readily
observable changes in land use are involved.
Setting a baseline and determining addition-
ality may be more difficult.
• Agricultural soil carbon sequestration appears
to have low leakage but may require significant
site-specific data to determine a baseline and
additionality and monitor project activities. Risk
of reversal from increased tillage is moderate to
high and may require site-specific data to assess.
• Agricultural CH4 and N2O mitigation options
and biofuels appear to have low leakage and
may have a low likelihood of reversal. Some
options (e.g., CH4 capture from manure man-
agement and biofuels) in general appear to be
readily monitorable and likely to be additional,
while others (e.g., soil N2O mitigation options)
may be more challenging to evaluate for these
issues.
• Biofuel offsets, though a relatively high-cost
option in the economic analyses above, have a
number of implementation advantages in that
they are relatively easy to measure, monitor,
and verify; highly additional under current
energy market conditions; and have low
reversal risk.
Taken together, it is interesting to observe that
some of the lower cost mitigation options found
in the economic analyses (e.g., forest management
and agricultural soil carbon sequestration) may
have implementation challenges, in contrast to
options such as biofuels implementation and
afforestation, which have higher opportunity
costs (in the economic analysis) but possibly lower
implementation transaction costs.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
6-13
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
Table 6-5: Implementation Issues for Selected Activities and Projects: Leakage Estimates from
FASOMGHG and MMV
Activity
Leakage Potential
(and Estimates)
MMV Difficulty
Afforestation
Moderate
U.S. average: 28%
Regions: 18-42%a
Relatively easy to measure, monitor, and verify forest
establishment. Measuring carbon is relatively
straightforward for above-ground carbon, less so
for below-ground carbon. Models can be used instead
of direct measurement if program allows.
Forest management
Likely some leakage
through reduced
afforestation
No separate estimates
available
Moderate to difficult to measure, monitor, and verify
specific management actions attributable to a project.
Measuring carbon in established stands is not
exceedingly difficult, but tying the change in carbon
to specific practices may be.
Agricultural soil Low
carbon sequestration 6%
Easy-moderate to measure, monitor, and verify across
adopting practices.
Moderate-difficult to directly estimate carbon consequences
across the landscape. Models can be used instead of direct
measurement if program allows.
Agricultural CH4 Low
and N2O mitigation NA
Easy (e.g., for manure management CH4 tied to
electricity-generating systems), difficult for dispersed
emissions (e.g., enteric fermentation at the herd level).
Biofuel offsets
Low
Easily tied to the biofuel market transactions.
Results from five regions in Murray et al. (2004) reported above.
Per-Acre Payments for Carbon
Sequestration to Address Measurement
GHG mitigation activity could be designed to
economize on transaction costs, particularly MMV
costs. The incentive approaches evaluated thus far
have paid for GHG mitigation on a dollar-per-
tonne basis. An alternative is for payments to be
based on a per-unit area (acre) tied to the adoption
of a specific mitigation practice. This approach is
similar to a number of land-based conservation
programs in the United States, such as the CRP
and The Environmental Quality Incentives Pro-
gram (EQIP). This approach may economize on
transaction costs because it relies on simple
verification that the land-use change has occurred
on the land in question, rather than quantification
of the GHG tonnes that have been mitigated. The
per-acre versus per-tonne issue is commonly
referred to as "practice versus performance
payments."
Scenario Description
Two of the carbon sequestration options consid-
ered thus far—afforestation and agricultural soil
carbon sequestration (tillage change)—are eval-
uated because they represent the dominant miti-
gation activities at medium-high and low GHG
prices, respectively, and they are distinct activities
that can be tracked relatively easily at the per-acre
level. Other activities may be more difficult to pay
for on a per-acre basis, because they are not space
extensive (e.g., CH4 and N2O mitigation activities
assessed in Chapter 4).
Per-acre results are evaluated against the targeted
$15/tonne CO2 payment scenario presented in
Chapter 5 (i.e., the situation under which the
selected activity—and only the selected activity—
6-14
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
receives payments at a rate of $15 per tonne). In the
per-acre payment case, the activity and only the
activity will receive payments of $100 and $15 per
acre per year for the entire 10-decade simulation
period for afforestation and tillage change activi-
ties, respectively. These per-acre values were
selected because they roughly reflect the equiva-
lent per-unit area payments of $15/tonne for
representative sequestration rates for the two
activities (about 6 to 71 CO2 per year for afforesta-
tion and 11 CO2 per year for tillage change).3
Two types of per-acre payment approaches are
evaluated for each activity:
• Uniform—any and all acres within the United
States that adopt the practice receive the same
Table 6-6: Qualitative Consideration of Implementation Issues for Selected Activities and Projects:
Baselines, Additionality, and Reversal Risk
Activity
Baseline Setting
Feasibility
Potential for
Additionality
Reversal Risk of GHG
Benefits (Permanence)
Afforestation
Agricultural CH4
and N2O mitigation
Biofuel offsets
Credible baseline at
adequate spatial and
temporal resolution is
likely. Involves observable
land-use change.
High in most places
within United States,
unless locally high
tree-planting rates.
Moderate if timber or land
prices change or natural
disturbances (fire, pests).
Forest
management
Protection
(avoided
deforestation)
Agricultural
soil carbon
sequestration
Difficult to observe
practices with remotely
sensed data. Includes
many practices varying
by forest type, etc.
Likely to require baseline
deforestation rates by
forest type and region,
projected into future.
Involves observable
land-use change.
Need data on continuous
tillage practices and rates
of alternative tillage
adoption.
Likely need to
demonstrate
introduction of
alternative practices.
Likely high if new
protection status
is conveyed or high
deforestation rates;
low, if not.
High if conventional
tillage persists into
future; low otherwise.
Moderate if timber or land
prices change or natural
disturbances (fire, pests).
Low if legal protection
and it is enforced. High
if susceptible to wildfire,
has uncertain legal status,
major commodity price
changes, etc.
Moderate-high: potential
seasonal tillage change
(weed control); or change
in crops or tillage practices
in response to commodity
prices or programs.
Remote sensing not
useful. Need activity data
per unit of production.
If adequate data, likely
credible baseline.
Moderate-high.
Low. No carbon storage
subject to re-release
involved.
Similar to afforestation
and soil tillage options
but may require energy
sector data to determine
baseline demands for
biofuels.
High based on recent
market trends.
Low. Primary benefit does
not involve carbon storage
subject to re-release,
although response to
changing commodity prices
could affect soil carbon.
3 Note that the per-acre payment values were based on average carbon yields per acre nationwide but, as shown below, the
realized gains per acre will be lower than average because of the inefficient nature of the incentive payments that either do not
differentiate or differentiate imperfectly by carbon yield per acre.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
6-15
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
per-acre payment for changing practices ($100
for afforestation and $15 for tillage change).
• Productivity based—any given acre receives
one of five payment levels for each activity.
The payments are based on the relative carbon
productivity of the acre.4
By at least partly basing payments on carbon
productivity, the productivity-based per-acre
payments should operate more closely to per-
tonne payments than uniform payments do. The
productivity-based approach more closely follows
programs such as the CRP, which have graduated
payments for changes in land use and practices
based on site characteristics. In contrast, the
uniform payments should induce more inefficiency.
The results below bear this out.
Per-Acre Payments for Carbon Sequestered
through Afforestation
Results of the per-acre payments for afforestation
are presented in Table 6-7 and compared to the $15
per-tonne afforestation-only payment scenarios
from Chapter 5. The uniform $100 per-acre pay-
ment approach is substantially less efficient than
the per-tonne approach. On an annualized basis
over the projection period, the uniform per-acre
payments generated only about 30 percent as much
sequestration as payments on a per-tonne basis
(41.9 vs. 137.4 Tg CO2 Eq.). However, the value of
the payments is about 60 percent as much ($790
MM vs. $1.36 billion). For the year 2015, which is
the midpoint of the first decade of the simulation,
only about one-quarter the amount of carbon is
sequestered even though one-half as much acreage
is afforested. This demonstrates a critical short-
coming of uniform per-acre payments, namely,
that the payments are made without regard to the
biophysical sequestration potential of the site—
each afforested acre receives the same payment.
Therefore, tonnes sequestered on a low productiv-
ity site are more costly than tonnes sequestered on
a high productivity site, which is an economically
inefficient way to sequester a given amount
of carbon.
Table 6-7 shows how modifying the payments
based on site productivity can improve the
effectiveness of the per-acre payment approach.
Productivity-based payments generate about 70
percent more carbon (annualized) than the
uniform payments, although the cost of the pay-
ments rises by only about one-third. In the first
decade (proxied by the 2015 results), the amount of
carbon sequestered matches that in the dollar-per-
tonne payment scenario. However, when compared
Table 6-7: Per-Acre vs. Per-Tonne Payment Approaches for Afforestation: 2015 and 2010-2110
Annualized
Payment Scenario
$15/t
C02 Eq.
$100/Acre
Uniform
$100/Acre
Productivity Based
Year 2015
GHG mitigated (Tg CO2 Eq. per year)
Net afforestation (MM acres)
Over 2010-2110 projection period (annualized)
88.8
10.1
23.5
5.1
89.9
11.3
GHG mitigated through afforestation
(Tg CO2 Eq. per year)
Value of GHG payments (billion $ per year)
137.4
$1.36
41.9
$0.79
68.6
$1.06
4 Candidate acres are ordered by carbon productivity and divided into quintiles. The middle quintile received the default value
payment ($15/acre for tillage change or $100/acre for afforestation), the top two quintiles received higher per-acre payments,
and the lowest two quintiles received lower per-acre payments. Payments were based on relative carbon productivity, yielding a
payment range of $5 to $16 per acre for tillage change and $65 to $130 per acre for afforestation.
6-16
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
to the per-tonne results over the entire projection
period, the productivity-based payment approach—
although superior to the uniform payment approach—
is still less efficient than the per-tonne approach in
that it generates only half as much carbon on an
annualized basis at a cost that is only about 22
percent lower. A payment approach that has more
than the five differentiated payments employed
here, however, would operate even more closely to
the per-tonne approach.
Changing the nature of the payments also changes
the regional distribution of afforestation responses
(see Figure 6-2). Under all payment approaches,
the South-Central region has the largest afforesta-
tion response (over 70 percent of the national
total); however, the uniform payment approach
shifts some of the South-Central's afforestation
carbon share to other regions, notably the Rocky
Mountains. Again, this reflects the change in
emphasis from paying for the highest carbon-
yielding afforestation to paying for any afforesta-
tion at the same amount. The Rocky Mountains
region's biophysical sequestration yield is less than
the South-Central region's but receives the same
payment and therefore comprises a larger share of
the program under uniform payments than under
per-tonne or distributed payments.
Per-Acre Payments for Agricultural Soil
Carbon Sequestered through Changes
in Tillage
Similar patterns emerge when comparing the
per-tonne and per-acre payment approaches for
agricultural soil carbon sequestration (see Table
6-8). As with afforestation, the uniform per-acre
payment approach is substantially less efficient
than the per-tonne or productivity-based payment
approach. The uniform payments cost more than
half as much as the per-tonne payments but yield
only about one-fifth as much carbon. This result is
similar to the findings of Antle et al. (2003), who
find that per-acre contracts for soil carbon seques-
tration are up to five times as expensive as per-
tonne contracts. As with afforestation, the ineffi-
ciency situation is partly remedied with the
introduction of productivity-based payments,
which generate more than half the amount of
carbon at about 85 percent of the cost of the per-
tonne approach.
The main factor underlying the inefficiency of
uniform payments is found by looking at the
distribution of tillage practices in the first decade
(2015). The primary response under uniform
payments is the adoption of conservation tillage,
rather than the more substantial zero tillage
practice. Farmers are paid the same for either
practice and therefore adopt the less costly conser-
vation tillage, even though it does not sequester as
much carbon.
Figure 6 2: Regional Shares of Afforestation Carbon Sequestration by Payment Approach
Shares based on annualized mitigation, 2010 2100.
1.0
§.2 0.8 -
0.6 -
0.4 -
0.2 -
0.0
PNWE
PSW
• $15/tCO2Eq.
• $100/ac Uniform
D$100/ac Productivity-Based
RM
Year
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
6-17
-------
CHAPTER 6 • IMPLICATIONS OF MITIGATION VIA SELECTED ACTIVITIES
The regional distribution of agricultural soil
carbon sequestration is also moderately affected
by the payment approach (see Figure 6-3). Moving
from per-tonne to a uniform per-acre payment, the
regional shares shift some from the Corn Belt and
Northern Plains to the Lake States and South-
Central regions. Switching to productivity-based
per-acre payments would restore the regional
shares to a pattern roughly the same as the per-
tonne payments.
Table 6-8: Agricultural Soil Carbon Sequestration Payment Approaches: 2015 and 2010-2110
Annualized
Payment Scenario
$15/t
C02 Eq.
$100/Acre
Uniform
$100/Acre
Productivity-Based
Year 2015
GHG mitigated (Tg CO2 per year)
Conservation tillage (MM acres)
Zero tillage (MM acres)
Over 2010-2110 projection period
GHG mitigated through tillage change
(Tg CO2, annualized)
Value of GHG payments (billion $, annualized)
Mitigation delivery efficiency
190.9
2.9
169.4
154.2
$1.61
41.9
119.5
60.1
33.7
$0.90
127.7
0.2
192.1
81.7
$1.36
Figure 6 3: Regional Shares of Agricultural Soil Carbon Sequestration by Payment Approach
NP
LS
NE
RM
SC
SE
l$15/tCO2Eq.
PNWE PSW
Region
$15/acre: Uniform D$15/acre: Productivity-Based
SW
6-18
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 7
Non-GHG Environmental
Co-effects of Mitigation
Chapter 7 Summary
Changes in land-use and management practices as a result of GHG mitigation actions can
produce non-GHG environmental co-effects. Wide-scale conversion of agricultural land to forest
may affect water quality, air quality, soil quality, and biodiversity. FASOMGHG predicts a net
increase in forestland of 5 million acres at the $15/t CO2 Eq. (or $55/t C Eq.) price and 58 million
acres at the $50/t CO2 Eq. (or $183/t C Eq.) price by the year 2055. All nonpoint source pollutant
loadings to national waterways modeled in FASOMGHG, except pesticides, are predicted to decline
from the baseline amounts under all GHG prices. Pesticides increase slightly under the low GHG
prices but decline under the higher prices. Even at low GHG prices, these reductions in nonpoint
source pollutant loadings may improve national and regional water quality, though effects would
likely vary substantially across regions. Co-effects of GHG mitigation on biodiversity (not modeled in
this analysis) may be both positive and negative. The net impact will depend on the baseline land
cover and type of cover to which it is converted in response to GHG incentives.
This report mainly focuses on quantifying
and evaluating the mitigation potential
for net GHG emission reductions through
forestry and agricultural activities. However,
the large-scale changes in land use and land
management practices projected in a number of
the mitigation scenarios could have a substantial
impact on resource flows in other (non-GHG)
aspects of environmental quality. GHG mitigation
co-effects in the forest and agriculture sectors
include changes in water quality, air quality, soil
quality, biodiversity, and aesthetics (McCarthy
et al. 2001). Therefore, assessing the net societal
effects of GHG mitigation will depend on more
inclusive analysis that captures a range of expected
effects within and across different impact catego-
ries (Elbakidze and McCarl 2004).
This chapter broadens the scope of the assessment
by examining some key ancillary land-use and
environmental effects that result from the forestry
and agricultural activities and analytical scenarios
described earlier. This report focuses on GHG
effects as the primary objective, so the non-
GHG environmental effects are reported here as
ancillary. Conversely, many existing land-based
programs are designed to attain non-GHG envi-
ronmental objectives (e.g., erosion control, reduced
nonpoint agricultural runoff, habitat preservation)
but also may have GHG consequences. In that
regard, GHG flows could be viewed as a co-effect
of those programs. While assessing the general
environmental effects of existing or proposed land
management programs and their concomittant
GHG benefits would be a way to estimate the
latter, this approach remains outside the scope
of this analysis.
One of the key changes projected by the FASOMGHG
model in most of the GHG mitigation scenarios is
large-scale adjustments in land use and land
management. As noted in Chapter 4, land tends
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
7-1
-------
CHAPTER? • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
to convert from agriculture to forests and biofuels
in response to GHG price incentives, particularly
under higher GHG prices. Underlying this general
trend are numerous adjustments across the major
land uses, namely cropland, timberland, pasture-
land, and land devoted to biofuels. For instance, at
higher GHG prices, biofuels play an important role
in GHG mitigation, and biofuel production uses
substantial land area.
To get a sense for the overall adjustments projected
by FASOMGHG, land uses are compared for
the baseline, $15, and $50 constant GHG price
scenarios for 2015 and 2055. (The $50 price is used
here to evaluate the effect of higher prices on
stimulating biofuel penetration, which is minimal
at lower prices.) Under the baseline, crop and
timberland use declines, while pastureland use
increases. For the two GHG price scenarios, land
use initially shifts heavily toward forests in 2015,
as expected. For the $15 per tonne CO2 scenario,
timberland area increases 19 million acres, and
for the $50 per tonne scenario, timberland area
increases by 97 million acres by 2015. By 2055,
however, much of this additional forest has
converted out of timberland into other uses. Net
timberland gain in 2055 for the $15 per tonne
scenario is only about 5 million acres, and for the
$50 per tonne scenario, it is around 58 million acres.
The results in Chapter 4 show that, as GHG prices
rise, biofuels become a more important part of the
future GHG mitigation portfolio. Table 7-1 illus-
trates the implications of that adjustment for land
use. Large areas of land, 42 million acres, are
ultimately devoted to biofuel production in the $50
per tonne CO2 Eq. GHG price scenarios by 2055.
Thus, although cropland and pastureland both
decline relative to the baseline, this land converts
to biofuel and forest uses.
Regional Distribution of Land Uses
Land-use changes projected to occur in response
to GHG price scenarios are not evenly distributed.
Figures 7-1 and 7-2 show the proportion of land in
each region devoted to different land uses in 2015
and 2055 under the baseline scenario and the $15
and $50 constant GHG price scenarios. Three
interesting trends emerge.
Table 7-1: Land Use under the Baseline, $15, and $50 (Constant) GHG Price Scenarios: 2015 and 2055
Quantities are in million acres.
Land Use
Baseline
GHG Price Scenario
($/t C02 Eq.)
$15
$50
2015
Cropland
Pastureland
Timberland
Biofuels
2055
Cropland
Pastureland
Timberland
Biofuels
332
384
333
0
241
448
303
0
325
381
352
0
229
444
308
4.5
296
370
430
1.4
161
409
361
42
Note: Land areas do not sum to the same value in each year because some uses are not included.
7-2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 7 • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
First, the proportion of land devoted to timber
increases in the eastern United States with GHG
prices. For higher GHG prices, the expansion of
timberland is substantial in regions with less
timberland initially, such as the Corn Belt. By
comparison, in the western United States, the
timberland proportion expands only slightly
relative to the baseline. Most of this expansion
Figure 7 1a: Land Use Allocation by Eastern U.S. Regions in 2015: Baseline and the $15 and $50
Constant GHG Price Scenarios
• Pastureland
• Cropland
D Biofuels
D Timberland
CO
Notes: NE = Northeast; LS = Lake States; CB = Corn Belt; SE = Southeast
Figure 7 1b: Land Use Allocation by Eastern U.S. Regions in 2055: Baseline and the $15 and $50
Constant GHG Price Scenarios
• Pastureland
• Cropland
D Biofuels
D Timberland
LU
CO
CO
Notes: NE = Northeast; LS = Lake States; CB = Corn Belt; SE = Southeast
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
7-3
-------
CHAPTER? • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
occurs at the $15 GHG price, while for the larger
$50 GHG price, there is little additional timber-
land expansion compared to the $15 GHG price
scenario. These results generally make sense in
that regions that already have substantial forest
area (e.g., the Northeast) or regions that have
few productive sites remaining for forests (i.e.,
many western regions) cannot substantially
Figure 7 2a: Land Use Allocation by Western U.S. Regions in 2015:
Constant GHG Price Scenarios
o 100%
Baseline and the $15 and $50
• Pastureland
• Cropland
D Biofuels
DTimberland
Notes: GP = Great Plains; SW = Southwest; RM = Rocky Mountains; PSW = Pacific Southwest; PNWE = Pacific Northwest,
East Side of Cascades (Pacific Northwest West Side of Cascades is not shown due to a lack of data.)
Figure 7 2b: Land Use Allocation by Western U.S. Regions in 2055: Baseline and the $15 and $50
Constant GHG Price Scenarios
o
DC
_c
o
(0
D
o
o
O)
2
8.
100% -
90% -
80% -
70% -
60% -
50% -
40% -
30% -
20% -
10% -
0%
to
CD
LO O
-i- LO
-------
CHAPTER 7 • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
increase timberland area with low or high
GHG prices.
Second, cropland area declines in all regions over
time under both GHG price scenarios, except in
the Southwest (SW). There are fewer alternative
uses for cropland in the Southwest region (i.e.,
fewer opportunities to plant trees and/or biomass
crops) where more cropland is irrigated. Irrigation
also makes less sense for alternatives such as
biofuels or timber production.
Third, biofuels become a more important compo-
nent of mitigation as GHG prices rise. Under the
$15/t CO2 Eq. constant GHG price scenario, only
land in the Northeast is devoted to biofuels. Under
a GHG price of $50 per tonne, however, over 40
million acres could be devoted to production of
biofuels nationally by 2055. Regionally, all of this
biofuel production occurs in the eastern United
States (Figures 7-la,b), since U.S. biofuel crops
generally are rainfed and require fairly productive
sites to be profitable with carbon prices. In most
regions, the increases in biofuel production occur
on cropland and pastureland, although in the Corn
Belt, biofuel production occurs to some extent
through conversion of timberland.
Timberland Management Intensity
Substantial changes in the intensity of forest
management are underway in the United States,
both in the baseline and in the mitigation cases.
The forest industry historically focused on methods
to extract large, old-growth trees in clear cuts up
to the mid-twentieth century. Methods to establish
and manage plantations began in earnest in the
1960s, and these efforts continue today.
The success of plantations and recent emphasis on
other, noncommercial values of forests has shifted
the focus in the last 20 years away from extracting
old-growth through large-scale clear-cutting. The
industry has shifted toward extracting smaller
trees from fast-growing plantations and using
alternative, less-intensive methods to extract
timber from natural, second-growth stands with
minimal forest damage.
The GHG mitigation scenarios explored in this
study may influence trends in forest extraction
(e.g., the intensification of plantation areas to
generate more carbon sequestration). FASOMGHG
model results suggest that GHG prices increase
timberland management intensity to enhance
carbon sequestration, via practices such as addi-
tional fertilizers to increase forest growth and
thinning operations undertaken to enhance yield.
Recent evidence from studies in the southern
United States suggests that nitrogen fertilizing,
chemical suppression of competition, and other
management intensifications can increase biomass
on sites from 6 to 20 percent (Siry 2002). With
carbon valued for GHG mitigation purposes, the
incentives for more intensive management could
be heightened.
Agricultural Nonpoint Pollutant Runoff
One of the most important environmental issues
facing agriculture in the United States is its contri-
bution, along with forest management and urban
development, to nonpoint source water pollution.
Nonpoint sources, particularly agriculture, are
considered to be the leading source of water
quality impairment in U.S. rivers, lakes, and
streams (EPA 2000). Siltation, nutrient runoff
(such as nitrogen and phosphorous), and pesti-
cides are the primary nonpoint water pollutants
from agriculture.
This section of the report focuses on four of the
most important runoff components from agricul-
ture: nitrogen, phosphorous, sediments, and
pesticides. Individual estimates of inputs or
loadings of these pollutants are shown for several
GHG price scenarios. For nitrogen and phospho-
rous, loadings are estimated using algorithms
from the EPIC model (Williams et al. 1989) imbed-
ded in FASOMGHG. For soil erosion, the outputs
are total soil erosion, based on the Modified
Universal Soil Loss Equation (MUSLE). It is not
possible here to quantify direct pesticide loadings
(field outputs). Therefore, changes in pesticide use
are presented to approximate loadings potential.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
7-5
-------
CHAPTER? • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
The substantial changes in land use and manage-
ment projected under some of the GHG mitigation
scenarios in Chapter 4 suggest there could be large
potential changes in water quality. First, there is
potential to reduce nonpoint source pollution
through land-use change, such as shifting land out
of agriculture and into forests, and establishing
perennial biofuel cover. Both forestry and biofuel
production typically use fewer inputs and produce
fewer pollutants than traditional crop agriculture.
Management inputs (chemical and mechanical) in
forestry are applied less frequently and less inten-
sively than in agriculture. There is less experience
in and information on pollutants arising from
biofuel production. The FASOMGHG model, how-
ever, does include nutrient and pesticide require-
ments as part of the production set for biofuels.
Second, changes in the management of agricul-
tural land could alter the magnitude and quality
of farm runoff. Adoption of conservation tillage
was originally developed to reduce soil erosion;
thus, adoption of conservation tillage to increase
soil carbon should reduce sediment lodgings from
soil erosion over time. Because phosphorous is
typically attached to soil particles, reductions in
soil erosion should also reduce phosphorous
entering rivers and streams.
The potential effect of conservation tillage on
nitrogen and pesticide runoff, however, is less
clear. Pesticide use often increases with the
adoption of conservation tillage (because of the
need for greater weed and other pest control),
and conservation tillage reduces yield for certain
important crops, such as corn. Consequently,
farmers may adjust by adopting more intensive
nitrogen and pesticide applications when they
adopt conservation tillage. Agricultural soil
management practices to mitigate N2O emissions
by reducing fertilizer use also have the joint
benefit of reducing nitrogen loadings.
The rest of this section looks more carefully at
the estimates provided by FASOMGHG for soil
erosion, phosphorous, nitrogen, and pesticides.
Each of the variables is evaluated relative to its
projected baseline level, normalized to a value of
100 for the purpose of cross-pollutant comparisons
over time, and across the range of constant GHG
price levels evaluated in Chapter 4.
Adoption of reduced tillage practices induced
by the GHG prices reduces soil erosion (Figure
7-3). Soil erosion reductions occur relatively
quickly, due mainly to rapid adoption of tillage
change and shifts in land from agriculture to
Figure 7 3: Soil Erosion Index over Time by (Constant) GHG Price Scenario (Baseline = 100)
120
8
V
TO
T3
100
80 -
2010
2020
2030
2040
2050
2060
Year
7-6
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 7 • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
forestry (i.e., over the first 10 to 20 years of the
model run). Over time, erosion levels gravitate
slightly back toward baseline levels. But these
erosion reductions produce annual benefits,
implying continuing improvements in water
quality over time. Baseline levels of erosion are
also declining over time, so that all of the paths
shown in Figure 7-3 represent net reductions in
erosion relative to today.
Estimated phosphorous loadings decline with
the introduction of GHG prices (Figure 7-4).
This decline is roughly proportional to the reduc-
tions in erosion, because phosphorous is attached
to soil particles. For higher GHG prices in the
range of $15 to $50, the reductions in loadings in
the initial period are roughly similar, suggesting
the maximum reduction in phosphorous may be
around 40 percent. In many cases, loadings begin
moving back toward baseline levels over time as
farmers increase inputs per hectare to make up for
yield losses associated with conversion to conser-
vation tillage. Loadings remain lower than base-
line levels in total, because overall cropland areas
tend to decline with GHG pricing.
Estimated nitrogen loadings decline in all
scenarios (Figure 7-5). These reductions, as a
percentage of baseline loadings, are smaller propor-
tionally than those for phosphorous and erosion.
The initial reduction ranges from 5 to 21 percent
under the GHG price scenarios considered. For the
lower GHG prices, reductions in nitrogen loadings
initially are relatively small, and loadings move
back toward baseline levels over time. For the
higher GHG prices (>$15 per tonne CO2), reduc-
tions in loadings are larger initially, but, after a
while, they begin to rise back toward baseline
levels.
The increase in nitrogen applications is in response
both to lower crop yields associated with conserva-
tion tillage and to higher crop prices. Under the
higher price scenarios, farmers in the FASOMGHG
model are shown to intensify the use of nitrogen
to increase overall production of crops on land
that remains in agriculture, and that increase
eventually leads to increased loadings over time
but still below baseline levels.
Pesticide applications increase relative to the
baseline for lower GHG prices (Figure 7-6),
as land shifts into conservation and zero-tillage
practices. With reduced tillage, farmers often
increase pesticide use to control for weeds, pests,
and other competition in lieu of mechanical
control through conventional tillage practices.
These increases result in greater overall pesticide
Figure 7 4: Phosphorous Loading Index over Time by (Constant) GHG Price Scenario (Baseline = 100)
120
20
2010 2020 2030 2040
Year
2050
2060
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
7-7
-------
CHAPTER? • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
releases under the low-price GHG scenarios. As
GHG prices rise, however, more land is converted
from agriculture to forestry and biofuels, and
aggregate pesticide applications and runoff are
projected to decline.
Changes in Agricultural Runoff and Water
Quality—Results from a Separate Case Study
Measuring the impacts of these nonpoint source
pollution outputs on ambient water quality levels
requires additional modeling. The relationship
between nutrient or soil runoff and water quality
is a complex one, and linking the loading results
described above to environmental outcomes is
difficult. The actual effects of changes in agricul-
tural runoff on water quality will depend on numer-
ous factors, including existing loads, assimilative
capacity, routing of the pollutants through the river
and stream network, and nutrient processes in the
water (including nutrient limitations), all of which
vary substantially from watershed to watershed.
Figure 7 5: Nitrogen Runoff Index over Time by (Constant) GHG Price Scenario (Baseline = 100)
120
8
Baseline
$1
$5
$15
$30
$50
20
2010 2020 2030 2040
Year
2050
2060
Figure 7 6: Pesticide Index over Time by (Constant) GHG Price Scenario (Baseline = 100)
120
8
Baseline
$1
$5
$15
$30
$50
20 H—
2010
2020 2030 2040 2050 2060
Year
7-8
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 7 • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
This section describes a previously conducted case
study to show water quality impacts associated
with GHG mitigation in agriculture, using a
related economic model linked to a water quality
model. Note that the case study is from a separate
analysis described in Pattanayak et al. (2005) and
is not directly a part of the GHG mitigation sce-
narios performed for this report. However, be-
cause the modeling framework and scenarios are
so similar between this study and Pattanayak et
al., it warrants further discussion here.
The case study linked ASMGHG (McCarl and
Schneider 2001), which is in essence the agricul-
tural component of the FASOMGHG model used
in this report, with the National Water Pollution
Control Assessment Model (NWPCAM), a model
developed by RTI International (Research Triangle
Institute) for EPA.
NWPCAM was used to estimate regional and
national water quality impacts of GHG mitigation
scenarios of $6.80 and $13.60 per tonne of CO2 ($25
and $50/t C, respectively), run through ASMGHG.
Similar to scenarios analyzed in this report, GHG
mitigation actions taken in ASMGHG include
afforestation, agricultural soil carbon sequestra-
tion through tillage changes, CH4 and N2O reduc-
tions through livestock and soil management
changes, and biofuel production.
One benefit of the NWPCAM model is that
it provides results on water quality outcomes
through a water quality index (WQI) that accounts
for the loading of different pollutants, as well as
the impacts of those pollutants in specific stream
segments. The WQI is on a scale from 0 to 100 and
was developed for NWPCAM based on work by
Vaughn (1986) and McClelland (1974).
A second benefit is that the NWPCAM model
projects stream impacts throughout the country,
allowing both for highly aggregate weighted
measures of water quality at the national and
regional levels, as well as for more spatially refined
results within regions.
Results for the $6.80 CO2 Eq. price scenario
showed, among other things, that CO2 makes
up most of the net GHG mitigation, a decline of
cropland production using conventional tillage, an
expansion of conservation tillage, and an increase
in afforestation of 5.8 million acres.
Figure 7-7 shows the water quality implications of
the $6.80 per tonne CO2 Eq. scenario distributed
across the continental United States. The water
quality changes reflect changes in loadings for all
GHG mitigation activities, except for afforestation
and livestock management. Note also that
ASMGHG and NWPCAM are both static models,
so the simulated water quality effects in Figure 7-7
are for a representative year (circa 2020, based on
data inputs to the models used). Dark blue indi-
cates substantial improvement in surface water
quality, light blue presents small to moderate
improvement, black spots indicate some water
quality degradation, and grey areas reflect no
appreciable change in water quality. For this
relatively low GHG price, the aggregate, national-
level surface WQI in NWPCAM increases by about
1.5 index points, which is a 2 percent improvement
in the WQI from its baseline levels. Effects are
primarily concentrated up and down the Missis-
sippi River Valley and west of the 100th meridian.
Nitrogen loadings into the Gulf of Mexico are
projected to decline by 144,000 tonnes per year
under this price scenario. This decline amounts to
about half of the national goal under the Watershed
Nutrient Task Force for solving the hypoxia problem
(Mississippi River/Gulf of Mexico Watershed
Nutrient Task Force 2001). These results are
generally consistent with those shown in modeling
of Gulf nitrogen loadings by Greenhalgh and
Sauer (2003), although that study used different
economic and biological models.
The changes vary across regions. Focusing on
the Corn Belt and Southeast regions, as well as the
nation as a whole, Table 7-2 shows the effects of
the $6.80 per tonne CO2 Eq. GHG price scenario
for the ASMGHG-NWPCAM simulation. The
national effects are consistent with the results for
the FASOMGHG model described above, although
total suspended solids increase nationally in the
case study. Loadings decline in the Corn Belt
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
7-9
-------
CHAPTER? - NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
region, because large areas of cropland are con-
verted to conservation tillage. In contrast loadings
increase in the Southeast mainly because there
are adjustments in the types of crops grown.
Despite this increase in loadings, the WQI for the
Southeast improves very slightly. As noted above,
the link between loadings and water quality out-
comes depends on numerous location factors. Even
within the Southeast, some regions experience
lower loadings and water quality improvements.
Figure 7-7; Changes in Water Quality from Soil Carbon Sequestration and Other Agricultural
Management Changes under $6.8 per Tonne CO2 Scenario in ca. 2020, using the
ASMGHG-NWPCAM Integrated Agriculture Water Quality Model
Change in Water
Quality Index
from Baseline
-40 to -1
0
1 to 6
7 to 100
Note: $6.80/t CO2 Eq. = $2571C Eq., the modeled price.
Source: Partanayak et al. (2005).
Table 7-2: Change in Pollutant Loadings for Selected Agricultural Pollutants and the WQI for the
$6.80 per tonne CO2 Eq. Scenario, using the ASMGHG-NWPCAM Model Integration
% Change in Pollutant Loading or WQI
Pollutant Loadings
Nitrogen
Phosphorous
Total suspended solids
Pesticides
WQI
Corn Beit
-2.4
-0.7
-2.4
-0.1
4.5
Southeast
1.3
0.3
0.2
1.7
0.7
National
-3.1
-2.0
0.5
0.9
2.0
7-10
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 7 • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
The improvements in these locations lead to
aggregate gains in water quality at the regional
level as loadings shift to areas that are less damag-
ing to water quality.
The Pattanayak et al. ASMGHG-NWPCAM study
suggests that, even for low GHG prices in the
range of $5 to $15 per tonne CO2, national-level
water quality will improve. At around $5 per tonne of
C02, this improvement could be around 2 to 3 percent
for the nation and over 4 percent for the Corn Belt,
relative to baseline WQI measures.1 The benefits
occur heavily in the middle part of the country, as
Figure 7-7 and Table 7-2 indicate, because the most
intensive agricultural crop production currently
occurs there.
Lastly, the reduction in nitrogen outputs specifi-
cally could benefit an emerging national water
quality issue, hypoxia in the Gulf of Mexico.
Implications for Biodiversity of GHG
Analysis of the impacts of GHG mitigation
programs on biodiversity has gained substantial
attention recently. Generally, increasing forest area
restores habitat for plant, aviary, and soil organ-
isms. It reduces forest fragmentation and connects
protected-area and habitat fragments by providing
corridors for seasonal or opportunist movement
of broad-ranging species with large home range
requirements (Wayburn et al. 2000; Franklin and
Forman 1987; Mladenoff et al. 1997; Peters and
Lovejoy 1992).
Huston and Marland (2003) and Gitay et al. (2002)
suggest that there could be both positive and
negative effects of terrestrial carbon sequestration
programs on biodiversity, depending on the
location. For instance, biofuel projects that remove
natural forest cover and replace it with monocul-
tural vegetation could reduce biodiversity locally.
Alternatively, restoring bottomland hardwoods
on agricultural lands in the southeastern United
States would return that part of the landscape
closer to its presettlement ecosystem and could
thereby increase biodiversity on a local and
regional scale. Huston and Marland (2003) and
Gitay et al. (2002), however, do not attempt to
quantify biodiversity impacts and mostly consider
local effects.
Assessing the net effects of GHG mitigation on
biodiversity is complicated. Plantinga and Wu
(2003) explore carbon management through
afforestation in Wisconsin and find that a scenario
that increases forest area by 25 percent would cost
$100 to $132 million to accomplish. Their findings
also indicate that this scenario would provide
additional consumptive and nonconsumptive wild-
life benefits of $61 million. Their study, however,
assumed that the new forests would be similar to
existing forests (i.e., landowners would not adjust
the species types to maximize carbon payments)
and that the forests would be managed in the same
fashion that forests are currently managed. This
result contrasts with other studies that argue that
carbon sequestration payments could lead to
suboptimal biodiversity outcomes (Caparros and
Jacquemont 2003).
Clearly, GHG mitigation activities can influence
biodiversity in positive and negative ways. The
remainder of this section focuses on results from
the FASOMGHG model scenarios that can provide
some insight into these potential impacts.
Several forest-sector trends in the FASOMGHG
results have potential implications for biodiversity.
One trend is that the GHG price scenarios imply
that more intensive management is aimed at
increasing the growing stock of timber and carbon.
Increasing the area of plantations is one such
intensification. Tree planting now occurs on more
than 2 million acres per year in the United States
(Haynes 2003), and planted pine occupies just over
30 million acres of the land base (almost one-fifth
of the U.S. South's timberland base). In the future,
the area of planted pine is expected to rise by a
factor of two-thirds by 2040, without considering
1 Regional WQI measures in NWPCAM are aggregated weighted averages of the WQI for each stream reach in the region,
weighted by the mile frontage of each reach.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
7-11
-------
CHAPTER? • NON-GHG ENVIRONMENTAL CO-EFFECTS OF MITIGATION
GHG prices (USDA Forest Service 2002). With
GHG pricing incentives, the area is projected to
expand even more.
If the additional plantations resulting from GHG
mitigation are planted on marginal or abandoned
agricultural land, these plantations likely will
improve biodiversity relative to current conditions.
If, instead, the plantations are substituted for
natural stands and managed in strict even-aged
rotations, these plantations could reduce bio-
diversity relative to the natural stands they
replace, as argued by Huston and Marland (2003).
Some afforestation of marginal cropland in the
Mississippi Alluvial Valley, however, uses a mix
of native bottomland hardwood species to enhance
biodiversity and restoration of native ecosystems
(e.g., Schlamadinger [2003]).
The overall area of timberland is expected to
increase under the GHG scenarios, suggesting that
new lands planted to trees will be planted on lands
that are currently agricultural. Conversion of
intensively cultivated agricultural lands to forest
cover, even a monocultural forest cover, is likely to
have positive—or at least nonadverse—effects on
biodiversity.2 Forest edge effects and the juxtaposi-
tion of different habitats, and corridors for species
movement are enhanced (Wayburn et al. 2000;
Peters and Lovejoy 1992).
Thus, it is likely that the new forests projected
by FASOMGHG will improve biodiversity relative
to maintaining agriculture. In addition, the
FASOMGHG model projects that forests will be
managed in longer rotations when GHG price
incentives are introduced. Longer rotations imply
less-intensive harvesting regimes (and less forest
and soil disturbance) and likely improved biodi-
versity. It is difficult to know with certainty which
of these effects will dominate—intensive monocul-
ture or expanding timberland area combined with
less-intensive management on some land. The
results of the scenarios explored in this report raise
questions, however, which should be addressed in
further research.
In addition to the forestry-biodiversity interaction,
other changes suggested by the results in this
report have biodiversity implications. As GHG
prices rise above $15 per tonne CO2, the results in
this report suggest that biomass energy becomes a
competitive option for mitigation, and the area of
land devoted to producing biomass crops expands.
Huston and Marland (2003) state several concerns
about the implications of using land for biomass
production and potential reductions in biodiversity
if this land involves removing natural timberland
cover, wetlands, or other natural areas. If land
devoted to biomass energy production involves
converting cropland to biomass, however, biodiver-
sity could increase.
Thus, the impacts of growth in biomass energy
production on biodiversity will depend on which
lands are converted for use. Given the aggregate
nature of the FASOMGHG model, it is difficult to
determine exactly what parcels of land will be
converted to biomass production, so this report
does not attempt to quantify these potential
impacts. However, biodiversity issues related to
biomass will become more important as carbon
prices rise, given the potential penetration of
biomass energy at the higher levels.
A final consideration relates to agricultural pro-
duction. The results in the model imply substantial
conversion to conservation and zero tillage, par-
ticularly at the lower GHG prices. Conservation
tillage improves the health and diversity of the soil
ecosystem (Lai et al. 1998) and would be expected
to improve soil quality indicators substantially at
the lower carbon prices. However, conservation
tillage often also involves increasing inputs, such
as chemical fertilizers and pesticides, which could
offset some of the environmental gains from
conservation tillage.
2 Conversion of native grasslands to tree plantations, however, could diminish unique prairie ecosystems (Gitay et al. 2002),
but this type of conversion is not expected to occur under the mitigation strategies analyzed in this report.
7-12
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTER 8
Summary of Insights on Key
GHG Mitigation Issues
This chapter concludes the report by showing
how the results of the analyses presented
in the previous chapters may have relevance
for key issues regarding GHG mitigation from the
forest and agriculture sectors.
Key Issues
Some key issues for GHG mitigation in forestry
and agriculture are described below.
Level of Mitigation Achieved. How much GHG
mitigation is sought from the forest and agriculture
sectors? This report evaluates forestry and agricul-
ture's potential to sequester carbon and reduce
GHG emissions under different scenarios. As
higher levels of mitigation are achieved, the
portfolio of activities expands, as does the cost
of mitigation.
Time Frame. When would the mitigation occur?
This is a particularly critical question for carbon
sequestration activities, which have complex time
dynamics. Sequestration can generate substantial
mitigation in the near to middle term (1 to 3
decades) but can decline after that because of
biophysical saturation and practice reversal. Some
alternatives such as biofuels have great technologi-
cal potential to mitigate GHGs immediately and
over the long term, but the infrastructure to
handle widespread adoption could take decades
to develop.
Comprehensiveness of Scope. Analytical results
show that nearly 2,000 Tg CO2 Eq. (or 2 billion
tonnes) per year of mitigation potential exists at
the highest-price scenario evaluated ($50/tonne
CO2 Eq.) if all private land, activities, and GHGs
are included. However, this rather large mitigation
potential can be reduced via criteria that narrow
the activities, GHGs, and time frames considered.
• Which activities and GHGs are included? Inclusion
could range from essentially all activities in
forestry and agriculture that have some meas-
urable GHG impact to a select few activities or
GHGs that are targeted for their cost-effective-
ness, desirable co-effects, or ease of monitoring.
• What land l>ase is included? The analysis in this
report has examined the mitigation potential
from all private lands in the conterminous
United States. But the scope could in principle
be larger or smaller than that. For instance,
public land can be managed to sequester carbon
and otherwise mitigate GHGs, but these actions
would presumably need to operate outside the
type of economic incentive-based system
evaluated in this report. Furthermore, programs
may focus on specific regions or states either for
economic or jurisdictional reasons.
Incentive Structure. The incentive structure
refers to the form that the GHG mitigation incen-
tives take and the appropriate incentive level for
a given mitigation quantity. Related questions
include the following:
• What are the units of exchange? For land-based
actions, a critical question is whether payments
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
8-1
-------
CHAPTERS • SUMMARY OF INSIGHTS ON KEY GHG MITIGATION ISSUES
are based on a per-tonne of CO2 Eq. or per-acre
basis. Although the latter is less costly to
measure, monitor, and verify (MMV), the
former tends to be much more efficient.
• What mechanisms can be used to induce mitigation
actions? In a purely market-based system,
mitigation incentives are determined by the
laws of supply and demand. In a government-
sponsored incentive program, compensation
levels may be administratively determined.
Accounting Requirements. How will GHG mitiga-
tion performance be measured? Related questions
include the following:
• Are GHG mitigation quantities measured at a specific
point in time, an average over some time period, or
cumulatively since the beginning of the program?
The amount attributed to an action can be
substantially affected by the completeness of
the accounting over time.
• Will adjustments be made to revise project-level
mitigation totals? Ideally, project quantification
should reflect net mitigation over time. This
suggests that adjustments may be necessary
to capture baseline emission or sequestration
levels that would have occurred without the
project, GHG effects induced outside the project
boundaries (leakage), and future carbon rever-
sal likely to occur after a project ends.
• Will non-GHG co-effects be included in mitigation
evaluations? The report has shown that miti-
gation actions may produce environmental
co-effects that could influence the desirability
of GHG mitigation strategies. If possible, should
these co-effects be quantified and thereby
modify the attractiveness of certain mitigation
options?
Infrastructure. What infrastructure or technical
assistance might be helpful or necessary for landowners
to realize potential mitigation opportunities? Stand-
ardized and widely available measurement,
monitoring, and verification guidelines and
methods, for example, may help landowners
overcome implementation barriers and engage
in mitigation activities.
Insights from Analyzed Results
With these fundamental issues in mind, the results
of the analyses throughout this report are used to
provide insights that could shed light on the
potential role of forestry and agriculture in GHG
mitigation. These insights are enumerated and
discussed below.
While national mitigation rates decline
over time (under constant price scenarios),
cumulative GHG mitigation steadily
increases.
Total national mitigation—under the scenario with
a constant GHG price of $15/t CO2 Eq. ($55/t C
Eq.)—is estimated to average almost 630 Tg CO2/yr
(172 Tg C) in the first decade, 655 Tg CO2/yr (179 Tg
C) by 2025, and decline to 86 Tg CO2/yr (23 Tg C) by
2055 (see Figure 8-1). The total range of constant
price scenarios evaluated is $1 to $50/t CO2 Eq.
($3.7 to $184/t C Eq.). A declining rate of annual
mitigation (i.e., occurring in a given year) over time
is the result of saturating carbon sequestration (to
a new equilibrium) in forestry and agriculture and
carbon losses after timber harvesting.
Figure 8 1: National GHG Mitigation at Three
Focus Dates by GHG Price: Average
Annual
2000
2015
2025
Year
2055
GHG Price
($/t C02 Eq.)
8-2
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • SUMMARY OF INSIGHTS ON KEY GHG MITIGATION ISSUES
Cumulative GHG mitigation (i.e., achieved in the
years up to a given year) for the $15/t CO2 Eq. and
other constant price scenarios steadily increases
(see Figure 8-2). This cumulative amount reaches
about 26,000 Tg CO2 (7,080 Tg C) by 2055. On an
annualized basis over 100 years, the $15/t CO2 Eq.
scenario generates 667 Tg CO2/yr (182 Tg C) in
GHG mitigation relative to the projected baseline.
Annualized results represent the net annualized
equivalent, or "annuity value," of all GHG mitiga-
tion over the entire 100-year period of analysis,
using a discount rate of 4 percent.
Identifying attractive activities may require
looking at a range of characteristics for
each option.
Each potential mitigation activity has a wide range
of characteristics that may make it more or less
desirable. Table 8-1 highlights some of the key
characteristics of each mitigation activity consid-
ered in this report: mitigation potential, regional-
ity, non-GHG co-effects, and reversal risk. Rever-
sal risk is particularly important if the action is
expected to be short-lived and liability provisions
are not in place to ensure that post-program
reversal is addressed. Other potentially important
considerations not included in this table (and not
explicitly modeled in this report) include issues
such as the difficulty of measuring, monitoring,
and verifying project-level GHG effects and
setting project baselines.
The quantity and timing of mitigation
can determine the selected activities.
Table 8-2 shows that modest mitigation quantities
(less than 300 Tg CO2 Eq. per year) may be
achieved in the near term, with activities that
primarily include agricultural soil carbon and
forest management, at less than $5/t CO2 Eq. More
ambitious levels require a different range of
activities (e.g., afforestation and biofuels) and
require $15 to 30/t CO2 Eq. and above. Long-term
mitigation requires permanent reductions in CO2
and non-CO2 emissions from agricultural practices
(achievable at a relatively low GHG price incen-
tive) and biofuel production. Biofuels are economi-
cally achievable only at the higher GHG prices and
in the longer run, primarily because of capacity
constraints on biofuel use in the short run.
Achieving a specific mitigation level within
a narrow time frame may shift emissions
to periods before and after the period of
interest.
The report examines scenarios in which an aver-
age annual mitigation quantity is set for Year 2025
(the midpoint of the decade 2020 to 2030), which is
Figure 8 2: Cumulative GHG Mitigation in Tg CO2 Eq.
$15/t CO Eq. Constant Real Price
30,000
D Biofuel offsets
D Crop management FF mitigation
D Ag CH4 and N2O mitigation
• Forest management
D Afforestation
• Agricultural soil C sequestration
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
8-3
-------
CHAPTERS • SUMMARY OF INSIGHTS ON KEY GHG MITIGATION ISSUES
Table 8-1: Characteristics of GHG Mitigation Activities
Activity
Afforestation
Forest
management
Agricultural
soil carbon
sequestration
Fossil fuel
mitigation from
crop production
Agricultural CH4
and N2O mitigation
GHG
Mitigation
Potential3
High
Moderate
Moderate-
low
Low
Low
Regions of
Emphasis
South-Central
and Corn Belt
South-Central
Southeast
Corn Belt
Lake States
Great Plains
South-Central
and Southwest
Corn Belt
Key Environmental Co-effects
Increases forest cover; improves water
quality; biodiversity effects either (+)
or (-) depending on characteristics
of new forests and ecosystem displaced
by new forests.
Enhances forest biological stock; longer
rotations can provide critical habitat.
Reduced erosion and nutrient runoff.
Small increase in pesticide use.
Negligible effects within forest
and agriculture sectors.
Air quality improvements from some
activities (e.g., manure management).
Reversal
Risk"
High
High
Moderate-
high
Low
Low
Biofuel offsets
Very high
Eastern
regions
Biodiversity effects depend on previous
land use
Low
a Mitigation potential refers to mitigation attained at the highest GHG prices evaluated in report scenarios.
b Individual activities or projects could have lower or higher reversal risk, depending on activity and site characteristics.
Table 8-2: Potential Implications of Mitigation Level and Time Frame
P Mitigation Quantity
(Tg CO2 Eq./year,
annualized, 2010-2100)
GHG Scenario
($/t C02 Eq.)
Primary Near-Term
Strategies
(By 2025)
Primary Long-Term
Strategies
(Beyond 2025)
Low (<300)
$1-$5
Agricultural soil
carbon sequestration
Forest management
Forest management
Emissions reduction
(CO2 and non-COJ
from agricultural activities
Medium (-300-1 ,400)
High (1,400+)
$5-$30 Afforestation
Forest management
$30+ Afforestation
Forest management
Forest management
Biofuels
Biofuels
Fossil fuel CO2 and
non-CO2 emission
reduction options
8-4
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • SUMMARY OF INSIGHTS ON KEY GHG MITIGATION ISSUES
then either maintained, increased, or dropped
after that period. Figure 8-3 (reproduced from
Figure 5-2) shows the results over time as the fixed
mitigation quantities vary.
The first unintended consequence is that the
absence of any fixed level for the first decade (2010
to 2020) means that GHG emissions could exceed
baseline levels, as producers substitute current
(unconstrained) emissions for future (constrained)
emissions. This is a form of temporal leakage and
is reflected in the initial negative values in Figure
8-3 and occurs under all variations of the scenario.
This situation ultimately reverses when the 2025
mitigation quantity is met. However, another nega-
tive consequence occurs when the initial 2025 level
is dropped thereafter (the second scenario in
Figure 8-3), which leads to a large reversal of the
carbon sequestered in the previous decades.
These negative consequences might be avoided if a
cumulative mitigation quantity from a base year
(e.g., 2010) onward is put in place instead of an
annual quantity for the future time period and if
the quantity is not dropped in the future.
Figure 8 3: Responses to Set Mitigation Quantities: Cumulative Mitigation to 2100
Quantities are Tg CO2 Eq. cumulative net emissions reduction below baseline.
Note: Scale varies for each graph, from 4,000 to 35,000 Tg COZ.
35,000
25,000
0-
m
CM
O 15,000
O)
5,000
O
O
O)
-5,000
D Biofuel offsets
D Crop management FF mitigation
• Forest management
D Ag CH4 and N2O mitigation
D Afforestation
• Agricultural soil C sequestration
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
(a) T-375-375
4,000
3,500 -
D Biofuel offsets
D Crop management FF mitigation
• Forest management
D Ag CH4 and N2O mitigation
D Afforestation
• Agricultural soil C sequestration
-500
2010 2020 2030 2040
2050 2060
Year
(c) T-375-0
2070 2080 2090 2100
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
8-5
-------
CHAPTERS • SUMMARY OF INSIGHTS ON KEY GHG MITIGATION ISSUES
Under scenarios of rising GHG payments,
forest and agriculture mitigation action
may be delayed.
Scenarios simulating a rising GHG price show an
increasing rate of GHG mitigation over the first
few decades. However, the constant price scenarios
show a declining rate of GHG mitigation over the
same time period. Three rising GHG price scena-
rios are evaluated: $3/t CO2 Eq. rising at 1.5 per-
cent and 4 percent/yr, respectively, and $20/t CO2
Eq. rising at $1.30/yr. The analyses in Chapter 4
found that, compared to constant-price scenarios,
rising prices can lead to delayed action (see Figure
8-4, reproduced from Figure 4-14 from Chapter 4).
The left side of Figure 8-4 shows the constant price
scenarios at different levels, and the right side of
the figure shows three rising-price scenarios.
Rising prices generally cause delayed mitigation.
The effect is most pronounced for the two scena-
rios with the higher rates of future price change.
The primary reason for the delay is the "one-shot"
nature of carbon sequestration activities. Under
rising prices, if mitigation activities occur too early,
more carbon will be sequestered at low prices in
the near term and less carbon at high prices in the
future. The economically optimal response, which
the FASOMGHG model generates by assuming
that landowners correctly know that prices will
rise at the given rate, is to delay sequestration
actions to take advantage of higher future prices.
GHG incentives reduce net emissions from
the forest and agriculture sectors below
baseline levels. If the incentives are strong
enough, the joint sectors could move from
a net emissions source to a sink.
The FASOMGHG baseline GHG projection for the
combined forest and agriculture sectors shows a
cumulative net source of emissions over time.1 The
mitigation scenarios (see Figure 8-5), however,
generate responses that either reduce the size of
the joint sector emissions source (at low GHG
prices) or even produce a net GHG sink (at high
GHG prices).
Figure 8 4: Constant Price Scenarios vs. Rising Price Scenarios and GHG Mitigation
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline for 2015 and 2055.
2,500
2015
2055
$1 $5 $15 $30 $50
$3 @ $3 @ $20 @
1.5% 4% $1.30
1 EPA's U.S. GHG inventory shows these combined sectors to be a net sink currently; however, the EPA inventory includes carbon
sequestration on public forest lands (an additional carbon sink), and FASOMGHG does not, thereby tipping the sectors' baseline
GHG balance to a net source in the model.
8-6
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • SUMMARY OF INSIGHTS ON KEY GHG MITIGATION ISSUES
Leakage potential from limiting included
mitigation activities may be largely confined
to the forest sector.
Model results in this report and in related research
show that leakage potential within the forest sector
can be moderate to high, depending on the activity
and region (see Chapter 6). If all GHG mitigation
activities in forestry and agriculture are included
in a comprehensive approach scenario, leakage is
negligible. Market effects elsewhere in the United
States are captured in the mitigation totals com-
puted by FASOMGHG.
However, if some forest activities and regions are
singled out for mitigation, some of the benefits
could be offset by emissions from other activities
and regions (see Table 8-3). The primary driver of
this leakage is the interaction between how much
land is devoted to forests, called the extensive
margin of forestry, and the intensity with which
forests are managed, called the intensive margin.
If only afforestation is included as a mitigation
activity, but not the management of existing
forests, the latter could suffer at the expense of the
former, leading to carbon losses from the decline
Figure 8 5: Cumulative Net Emissions/Sinks for Forestry and Agriculture: Comparison of Baseline
and Comprehensive Mitigation Scenarios at Constant Prices over Time
60,000
u -20,000 -
CM
8 -40,000 -
u>
"~ -60,000
-80,000 -
-100,000
-120,000
(+) values represent net sources of emissions
(-) values represent net sinks
Baseline
$1
$5
$15
$30
$50
2015 2025 2035 2045
2055
Year
2065 2075 2085 2095
Table 8-3: Leakage Estimates by Mitigation Activity at a GHG Price of $15/t CO2 Eq.
All quantities are on an annualized basis for the time period 2010-2100.
selected Mitigation Activities
National Average Leakage Rate (%)
Afforestation only
24.0
Afforestation + forest management
-2.8
Biofuels
0.2
Agricultural management
-0.1
Agricultural soil carbon
5.7
Note: Negative sign indicates beneficial leakage (i.e., the selected activity increases mitigation in the nonselected activities).
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
8-7
-------
CHAPTERS • SUMMARY OF INSIGHTS ON KEY GHG MITIGATION ISSUES
in management. However, if both afforestation and
forest management are given incentives, the
results suggest that this leakage incentive essen-
tially disappears (see Table 8-3).
The agricultural activities evaluated in this report
do not appear to be as prone to leakage as forestry
activities. Leakage estimates from the agricultural
options were found to be less than 6 percent of the
direct mitigation benefits. The reason for more
limited leakage effects in agriculture is that the
changes in agricultural practices do not have as
profound an impact on agricultural commodity
markets as the forest activities do on timber
markets.
Raising GHG mitigation levels in forestry
and agriculture can cause environmental
co-effects, both good and bad.
Large changes in land use and production can also
have a substantial impact on non-GHG environ-
mental outcomes in forestry and agriculture,
primarily because of the role of agricultural soil
carbon sequestration in the mitigation portfolio
at a fairly low GHG price scenario (e.g., $5/tonne
CO2 Eq.). Even such a low GHG price can induce
changes in tillage practices across many cropland
acres. These practice changes also reduce erosion
and nutrient runoff to waterways as a co-benefit
but can lead to a modest increase in pesticide
use as a co-cost (Figure 8-6). Other potential
environmental effects, such as biodiversity issues,
are not modeled in this report but are addressed in
Chapter 7.
Taking these environmental co-effects into consid-
eration could affect the relative attractiveness of
competing mitigation options. In general, a modest
GHG mitigation action will probably have negli-
gible effects on non-GHG outcomes within the
sectors. However, the more aggressive the mitiga-
tion action, the more likely that co-effects may
factor into the net benefits of GHG mitigation.
Payment method will determine efficiency
of mitigation activities.
Paying on a per-tonne CO2 Eq. basis is more
efficient than paying on a per-acre basis to gener-
ate additional GHG mitigation. Compared to the
scenario paying for afforestation only (at $15/t CO2
Eq.), paying for afforestation on a uniform $100
per-acre basis generates only 30 percent as much
additional carbon but requires 60 percent as much
in payments. Per-acre payments do not directly
vary with the biophysical potential of the site. The
inefficiency could be remedied somewhat by
adjusting per-acre payments based on land
productivity.
Figure 8 6: Nitrogen Runoff Index over Time by (Constant) GHG Price Scenario
120
100 <>
o
o
o
in
TO
E9.
x
o
T3
Baseline
$1
$5
$15
$30
$50
2020
2030 2040
Year
2050
2060
8-8
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
CHAPTERS • SUMMARY OF INSIGHTS ON KEY GHG MITIGATION ISSUES
If outreach is needed to deliver GHG mitiga-
tion, these efforts might focus in regions
with the largest mitigation potential.
As shown in Figure 8-7 (reproduced from Figure
4-11), the regional distribution of mitigation
opportunities is skewed toward the eastern United
States. Federal and other public lands are not
included in this analysis, thereby ignoring mitiga-
tion potential on those lands. However, public
lands management, if included, would clearly
elevate the role of the western United States in a
national strategy. On the remaining private lands,
however, the regional distribution does vary some
with the level of mitigation sought. At low levels of
mitigation and prices, the two South regions
(South-Central and Southeast), via forest manage-
ment, and two Midwest regions (Corn Belt and
Lake States), via agricultural soil carbon seques-
tration, are the focal regions and activities. As
prices rise and mitigation levels expand, farmers
in the South and Midwest may participate by
planting trees on agricultural land. If GHG incen-
tives are strong enough to induce biofuel produc-
tion, landowner participation could expand
beyond the Midwest and South to include the
Northeast region.
Figure 8 7: Total Forest and Agriculture GHG Mitigation by Region
Quantities are Tg CO2 Eq. per year net emissions reduction below baseline, annualized over the time
period 2010 2110.
500
400
<
Q.
m
ev
O
o
01
300
200
100
NE
SE
LS
CB
$30
$15 GHG Price
($/tC02Eq.)
Region
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
8-9
-------
CHAPTERS • SUMMARY OF INSIGHTS ON KEY GHG MITIGATION ISSUES
8-10 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
Adams, D.M., R.J. Alig, J.M. Callaway, B.A. McCarl, and S.M. Winnett (1996) "The Forest and
Agriculture Sector Optimization Model (FASOM): Model Structure and Policy Applications."
USDA Forest Service, Pacific Northwest Research Station, Portland, Oregon. Research Paper
PNW-RP-495. 60 p.
Adams, Darius. M., Ralph J. Alig, Bruce A. McCarl, John M. Callaway, and Steven. M. Winnett
(1999) "Minimum Cost Strategies for Sequestering Carbon in Forests." Land Economics 75(3):
360-74.
Adams, R.M., J.M. Callaway, C. Chang, D.M. Adams and B.A. McCarl (1993) "Sequestering
Carbon on Agricultural Land: Social Cost and Impacts on Timber Markets." Contemporary
Policy Issues (January): 76-87.
Adams, Darius M and Richard W. Haynes. (1996) The 1993 Timber Assessment Market Model:
Structure, Projections, and Policy Simulations. USDA Forest Service Gen. Tech. Report PNW-
GTR-368. Portland, Oregon, 58 p.
Alig, R., D. Adams, B. McCarl, J. Callaway, and S. Winnett (1997) "Assessing Effects of Mitiga-
tion Strategies for Global Climate Change with an Intertemporal Model of the U.S. Forest and
Agriculture Sectors." Critical Reviews in Environmental Science and Technology 27: S97-S111.
Alston J.M., and B.H. Hurd (1990) "Some Neglected Social Costs of Government Spending in
Farm Programs." American Journal of Agricultural Economics 72(1): 149-156.
Antle, J., S. Capalbo, S. Mooney, E. Elliott, and K. Paustian (2003) "Spatial Heterogeneity,
Contract Design, and the Efficiency of Carbon Sequestration Policies for Agriculture." Journal
of Environmental Economics and Management 46(2): 231-250.
Ball, E.V., J.P. Butault, and R. Nehring (2001) U.S. Agriculture, 1960-96: A Multilateral Comparison
of Total Factor Productivity ERS. Technical Bulletin No. 1895. U.S. Department of Agriculture,
Economic Research Service. 56 pp. Data for figure downloaded from .
Baumes, H. (1978) "A Partial Equilibrium Sector Model of U.S. Agriculture Open to Trade:
A Domestic Agricultural and Agricultural Trade Policy Analysis." Ph.D. thesis, Purdue
University.
Birch, T.W., and R.J. Moulton (1997) "Northern Forest Landowners: A Profile." National
Woodlands (January).
Birdsey, R.A. (1992) "Changes in Forest Carbon Storage from Increasing Forest Area and
Timber Growth." In: Forests and Global Change, Vol. 1: Opportunities for Increasing Forest Cover.
R.N. Sampson and D. Hair (eds.), pp. 23-39. (App. 2), American Forests, Washington, DC.
Birdsey, R.A. (1996) "Regional Estimates of Timber Volume and Forest Carbon for Fully
Stocked Timberland, Average Management After Final Clearcut Harvest." In: Forests and Global
Change: Vol. 2, Forest Management Opportunities for Mitigating Carbon Emissions, R.N. Sampson
and D. Hair (eds.), pp. 309-334, American Forests, Washington, DC.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE R-1
-------
REFERENCES
Blanco, ]., and C. Forner (2000) "Special Considerations Regarding the 'Expiring CERs'
Proposal." The Ministry of the Environment, Colombia, for presentation to the International
Forum on Enhancement of Japan's Private Sector's Re-afforestation Cooperation.
Brown, S. (2002) "Measuring Carbon in Forests: Current Status and Future Challenges."
Environmental Pollution 116: 363-372.
Brown, S., O. Masera, and J. Sathaye (2000) "Project-Based Activities." In: Land Use, Land Use
Change, and Forestry, R.T. Watson, I.R. Noble, B. Bolin, N.H. Ravindranath, D.J. Verardo, and
D.J. Dokken (eds.), Intergovernmental Panel on Climate Change, Cambridge University Press,
Cambridge, UK. 377 pp.
Burton, R.O. (1982) "Reduced Herbicide Availability: An Analysis of the Economic Impacts on
U.S. Agriculture." Ph.D. thesis, Purdue University.
Callaway, J.M., and B.A. McCarl (1996) "The Economic Consequences of Substituting Carbon
Payments for Crop Subsidies in U.S. Agriculture." Environmental and Resource Economics
7(1996): 15-43.
Caparros, A., and F. Jacquemont (2003) "Conflicts Between Biodiversity and Carbon Seques-
tration Programs: Economic and Legal Implications." Ecological Economics 46:143-157.
Chang, C.C., B.A. McCarl, J.W. Mjelde, and J.W. Richardson. 1992. "Sectoral Implications of
Farm Program Modifications." American Journal of Agricultural Economics 74(1992): 38-49.
Chomitz, K.M. (2000) "Evaluating Carbon Offsets from Forestry and Energy Projects: How
Do They Compare?" World Bank Policy Research Working Paper 2357, New York, p. 25.
Chomitz, K.M. (2002) "Baseline, Leakage and Measurement Issues: How Do Forestry and
Energy Projects Compare?" Climate Policy 2(1): 35-49.
Climate, Community & Biodiversity Alliance (CCBA) (2004) "Climate, Community & Biodiver-
sity Project Design Standards (Draft 1.0)." CCBA, Washington DC. .
Conservation Technology Information Center (CTIC) (1994) 1994 National Crop Residue Manage-
ment Survey. Conservation Technology Information Center, West Lafayette, Ind.
Consortium for Agricultural Soils Mitigation of Greenhouse Gases (CASMGS) (2003) Carbon
Measurement and Monitoring Forum. .
DeAngelo, B.J., F.C. de la Chesnaye, R.H. Beach, A. Sommer, and B.C. Murray (in press)
"Methane and Nitrous Oxide Mitigation in Agriculture." Energy Journal.
Donigian, A.S., Jr., A.S. Patwardhan, R.B. Jackson, IV, T.O. Barnwell, Jr., K.B. Weinrich, and
A.L. Rowell (1995) "Modeling the Impacts of Agricultural Management Practices on Soil
Carbon in the Central U.S." In: Soil Management and Greenhouse Effect. R. Lai, J. Kimble,
E. Levine, and B.A. Stewart (eds.) CRC Press, Inc., Boca Raton, FL. pp. 121-135.
R-2 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
REFERENCES
Dutschke, M. (2001) "Permanence of CDM Forests or Non-permanence of Land Use-Related
Carbon Credits?" Hamburgisches Welt-Wirtschafts-Archiv (HWWA) Discussion Paper #134.
.
Dutschke, M. (2002) "Fractions of Permanence-Squaring the Cycle of Sink Carbon Account-
ing." Mitigation and Adaptation Strategies for Global Change 7: 381-402.
Elbakidze, L., and B.A. McCarl (2004) "Sequestration Offsets versus Direct Emission Reduc-
tions: Consideration of Environmental Externalities." Mimeo. Department of Agricultural
Economics, Texas A&M University.
Eve, M.D., K. Paustian, R. Follett, and E.T. Elliot (2000) United States Submission on Land-Use,
Land-Use Change, and Forestry, August 1. U.S. Department of State, U.S. Submission to the
United Nations Framework Convention on Climate Change, August 1:46.
Fama, E.F. (1970) "Efficient Capital Markets." Journal of Finance 25: 383-417.
Feng, Hongli, Jinhua Zhao, and Catherine L. Kling (2001) "Carbon: The Next Big Cash Crop?"
Choices Second Quarter: 16-19.
Follett, R.F., J.M. Kimble, and R. Lai (2001) The Potential of U.S. Grazing Lands to Sequester Carbon
and Mitigate the Greenhouse Effect, Lewis Publishers, Boca Raton, FL.
Franklin, J.F., and R.T. Forman (1987) "Creating Landscape Patterns by Forest Cutting:
Ecological Consequences and Principles." Landscape Ecology 1: 5-18.
Gitay, Habiba, Avelino Suarez, Robert T. Watson, and David Jon Dokken (2002) "Climate
Change and Biodiversity." Technical Paper, United Nations Intergovernmental Panel on
Climate Change, New York, NY.
Greenhalgh, S., and A. Sauer (2003) "Awakening the Dead Zone: An Investment for Agricul-
ture, Water Quality, and Climate Change." Issue Brief. World Resources Institute, Washington,
DC.
Hamilton, S.A. (1985) "The Economic Effects of Ozone on U.S. Agriculture: A Sector Modeling
Approach." Ph.D. dissertation, Oregon State University.
Haq, Zia (2002) "Biomass for Electricity Generation." EIA Forecasts, Energy Information
Administration Modeling and Analysis Papers, .
Harmon, M.E., W.K. Ferrell, and J.F. Franklin (1990) "Effects on Carbon Storage of Conversion
of Old-Growth Forests to Young Forests." Science 247(4943): 699-702.
Haynes, Richard W (2003) An Analysis of the Timber Situation in the United States: 1952 to 2050.
General Technical Report PNW-GTR-560. U.S. Department of Agriculture, Forest Service,
Pacific Northwest Research Station, Portland, Oregon.
Heath, Linda S. (2000). "Assessing Carbon Sequestration on Public Timberlands in the Conter-
minous United States." In: Seventh Symposium on Systems Analysis in Forest Resources. Michael J.
Vasievich, J., Jeremy S. Fried, and Larry A. Leefers (eds.). p. 257-260. USDA Forest Service,
General Technical Report NC-GTR-205. 470 p.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE R-3
-------
REFERENCES
Herzog, H., K. Caldeira, and J. Reilly (2003) "An Issue of Permanence: Assessing the
Effectiveness of Temporary Carbon Storage." Climatic Change 59(3): 293-310
Hobbs, R, and N. Stoops (2002) "U.S. Census Bureau, Census 2000 Special Reports, Series
CENSR-4, Demographic Trends in the 20th Century." U.S. Government Printing Office,
Washington, DC. .
Huston, M.A. and G. Marland (2003) "Carbon Management and Biodiversity." Journal of
Environmental Management 67: 77-86.
Intergovernmental Panel on Climate Change (IPCC) (1996) "Revised 1996 Guidelines for
National Greenhouse Gas Inventories." J.T. Houghton, L.G. Meira Filho, B. Lim, K. Treanton,
I. Mamaty, Y. Bonduki, D.J. Griggs, and B.A. Callander (eds.). Cambridge University Press,
Cambridge, UK.
Intergovernmental Panel on Climate Change (IPCC) (2000) Land Use, Land Use Change, and
Forestry, Summary for Policy makers. Cambridge University Press, Cambridge, UK. 377 pp.
Intergovernmental Panel on Climate Change (IPCC) (2001) Climate Change 2001: Mitigation.
Cambridge University Press, Cambridge, UK.752 pp.
Johnson, D.E., H.W. Phetteplace, A.R Seidl, U.A. Schneider, and B.A. McCarl (2003a) "Manage-
ment Variations for U.S. Beef Production Systems: Effects on Greenhouse Gas Emissions and
Profitability." Presentation at 3rd International Methane & Nitrous Oxide Mitigation Confer-
ence, Beijing, China, December.
Johnson, D.E., H.W. Phetteplace, A.R Seidl, U.A. Schneider, and B.A. McCarl (2003b) "Selected
Variations in Management of U.S. Dairy Production Systems: Implications for Whole Farm
Greenhouse Gas Emissions and Economic Returns." Presentation at 3rd International
Methane & Nitrous Oxide Mitigation Conference, Beijing, China, December, 2003.
Johnson, K.N., and J.L. Scheurman (1977) "Techniques for Prescribing Optimal Timber
Harvest and Investment under Different Objectives: Discussion and Synthesis." Monograph
18, supplement to Forest Science. 23.
Johnson, M.G. (1995) "The Role of Soil Management in Sequestering Soil Carbon." In: Soil
Management and Greenhouse Effect. R. Lai, J. Kimble, E. Levine, and B.A. Stewart (eds.) CRC
Lewis Publishers, Boca Raton, pp. 351-363.
Kadyszewski, John (2001) Testimony before the Senate Agricultural Committee, March 29,
2001.
Kauppi, P., and R. Sedjo (2001) "Technological and Economic Potential of Options to Enhance,
Maintain, and Manage Biological Carbon Reservoirs and Geo-engineering." Ch. 4 in Climate
Change 2001: Mitigation. Cambridge University Press, Cambridge, UK. 752 pp.
Kern, J., and M. Johnson (1993) "Conservation Tillage Impacts on National Soil and
Atmospheric Carbon Levels." Soil Science Society of America Journal 57: 200-210.
Lai, R., J.M. Kimble, R.R Follett, and C.V. Cole (1998) The Potential of U.S. Cropland to Sequester
Carbon and Mitigate the Greenhouse Effect, Lewis Publishers, Boca Raton.
R-4 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
REFERENCES
Lee, H. (2002) "The Dynamic Role for Carbon sequestration by the U.S. Agricultural
and Forest Sectors in Greenhouse Gas Emission Mitigation." PhD Thesis, Department of
Agricultural Economics, Texas A&M University.
Lewandrowski, J., M. Peters, C. Jones, R. House, M. Sperow, M. Eve, and K. Paustian (2004)
Economics of Sequestering Carbon in the U.S. Agricultural Sector. Technical Bulletin Number
1909. Washington, DC: U.S. Department of Agriculture, Economic Research Service.
Lugo, A.E., and S. Brown (1986) "Steady State Terrestrial Ecosystems and the Global Carbon
Cycle." Vegetation 68: 83-90.
Marland, G., K. Fruit, and R. Sedjo (2001) "Accounting for Sequestered Carbon: The
Question of Permanence." Environmental Science and Policy 4: 259-268.
McCarl, B.A. (1982) "Cropping Activities in Agricultural Sector Models: A Methodological
Proposal." American Journal of Agricultural Economics 64(4): 768-772.
McCarl, B.A., and D. Gillig (2003) Notes on Biomass Assumptions in FASOMGHG.
Department of Agricultural Economics, Texas A&M University, College Station.
McCarl, B.A., and B.C. Murray (2002) "Harvesting the Greenhouse: Comparing Biological
Sequestration with Emissions Offsets." Working paper.
McCarl, B.A., and U.A. Schneider (2000) "U.S. Agriculture's Role in a Greenhouse Gas
Emission Mitigation World: An Economic Perspective." Review of Agricultural Economics
22(1): 134 159.
McCarl, B.A., and U.A. Schneider (2001) "Greenhouse Gas Mitigation in U.S. Agriculture
and Forestry." Science 294: 2481-82.
McCarthy, J.J., O.F. Canziani, N.A. Leary, D.J. Dokken, and K.S. White (eds.) (2001) Climate
Change 2001: Impacts, Adaptation & Vulnerability. Contribution of Working Group II to the
Third Assessment Report of the Intergovernmental Panel on Climate Change (IPCC). Cambridge
University Press, UK.
McClelland, N.I. (1974) Water Quality Index Application in the Kansas River Basin. Prepared for
U.S. Environmental Protection Agency—Region VII. EPA-907/9-74-001.
Mills, J., and J. Kincaid (1992) The Aggregate Timberland Assessment Model-ATLAS: A Compre-
hensive Timber Projection Model. General Technical Report PNW-GTR-28 1. USDA Forest
Service, Pacific Northwest Research Station. Portland, OR 160 pp.
Mississippi River/Gulf of Mexico Watershed Nutrient Task Force (2001) "Action Plan
for Reducing, Mitigating, and Controlling Hypoxia in the Northern Gulf of Mexico."
Washington, DC.
Mladenoff, D.J., R.G. Haight, A.P. Wydeven, and T.A. Sickley (1997) "Consequences of
Species Recovery in Altered Ecosystems: A Spatial Landscape Projection of Wolf Population
Recovery in the Northern Great Lakes Region." Bioscience 47: 21-31.
Mooney, S., J.M. Antle, S.M. Capalbo, and K. Paustian. 2004a. "Design and Costs of a
Measurement Protocol for Trades in Soil Carbon Credits." Canadian Journal of Agricultural
Economics. 52(3): 257 287.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE R-5
-------
REFERENCES
Mooney, S., J.M. Antle, S.M. Capalbo, and K. Paustian. 2004b. "Influence of Project Scale on
the Costs of Measuring Soil C Sequestration." Environmental Management 33(supplement 1):
S252-S263.
Moulton, R.J. and K.R. Richards (1990) Costs of Sequestering Carbon Through Tree Planting and
Forest Management in the United States. USDA Forest Service. General Technical Report GTR
WO-58. Washington, DC, 47p.
Moura Costa, P. (1996) Tropical Forestry Practices for Carbon Sequestration. In: Dipterocarp
Forest Ecosystems-Towards Sustainable Management. A. Schulte and D. Schone (eds.). World
Scientific, Singapore, pp. 308-334.
Murray, B.C. (2000) "Carbon Values, Reforestation, and 'Perverse' Incentives Under the
Kyoto Protocol: An Empirical Analysis." Mitigation and Adaptation Strategies for Global Change 5:
271-95.
Murray, B.C., B.A. McCarl, and H. Lee (2004) "Estimating Leakage from Forest Carbon
Sequestration Programs." Land Economics 80(1): 109-124.
Muth, John A. (1961) "Rational Expectations and the Theory of Price Movements." Econometrica
29(6): 315-35.
Parks, P.J. and I.W Hardie (1995) "Least Cost Carbon Reserves: Cost-Effective Subsidies to
Convert Marginal Agricultural Land to Forest." Land Economics 71:122-36.
Parton, WJ. (1996) "The CENTURY Model." In: Evaluation of Soil Organic Matter Models Using
Existing Long-Term Datasets, D.S. Powlson, P. Smith, and J.U. Smith (eds.), Springer-Verlag,
Berlin, Germany, pp. 283-293.
Pattanayak, Subhrendu K., Bruce A. McCarl, Allan J. Sommer, Brian C. Murray, Timothy
Bondelid, Dhazn Gillig, and Benjamin DeAngelo (2005) "Water Quality Co-effects of
Greenhouse Gas Mitigation in U.S. Agriculture." Climatic Change 71(3): 341-372.
Peters, R.L. and T.E. Lovejoy (1992) Global Warming and Biological Diversity. Yale University
Press, New Haven, CT.
Phillips, O.L., Y. Malhi, N. Higuchi, W.F. Laurance, P.V. Nunez, R.M. Vasquez, S.G. Laurance,
L.V. Ferriera, M. Stern, S. Brown, and J. Grace (1998) "Changes in the Carbon Balance of
Tropical Forest: Evidence from Long-Term Plots." Science 282: 439-442.
Phillips, O.L., Y. Malhi, B. Vinceti, T. Baker, S.L. Lewis, N. Higuchi, W.F. Laurance, P. Nunez
Vargas, R. Vasquez Martinez, S. Laurance, L.V. Ferreira, M. Stern, S. Brown, and J. Grace
(2002a) "Changes in Growth of Tropical Forests: Evaluating Potential Biases." Ecological
Applications 12(2): 576-587.
Phillips, O.L., R.V. Martinez, L. Arroyo, T. R. Baker, T. Killeen, S. L. Lewis, Y. Malhi,
A.M. Mendoza, D. Neill, P.N. Vargas, M. Alexiades, C. Ceron, A. Di Fiore, T. Erwin, A. Jardim,
W. Palacios, M. Saldias, and B. Vinceti (2002b) "Increasing Dominance of Large Lianas in
Amazonian Forests." Nature 418:770-774.
Plantinga, A.J., T. Mauldin, and D.J. Miller (1999) "An Econometric Analysis of the Costs of
Sequestering Carbon in forests." American Journal of Agricultural Economics 81(November):
812-24.
R-6 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
REFERENCES
Plantinga, A.J. and J. Wu (2003) "Co-Benefits from Carbon Sequestration in Forests:
Evaluating Reductions in Agricultural Externalities from an Afforestation Policy in
Wisconsin." Land Economics 79 (1): 74-85.
Richards, K. (1997) "Estimating Costs of Carbon Sequestration for a United States
Greenhouse Gas Policy." Report prepared for Charles River Associates.
Richards, Kenneth R. et al. (Forthcoming) "A Brief Overview of Carbon Sequestration
Economics and Policy." Environmental Management.
Richards, K., Moulton, R., and Birdsey, R. (1993) "Costs of Creating Carbon Sinks in the
U.S." Energy Conservation and Management 34(9-11): 905-912.
Richards K.R., and C. Stokes (2004). "A Review of Forest Carbon Sequestration Cost Studies:
A Dozen Years of Research." Climatic Change (68): 1-48.
Roningen, Vernon O., John Sullivan, and Praveen Dixit. (1991) Documentation of the Static
World Policy Simulation (SWOPSIM) Modeling Framework. Washington, DC, Economic
Research Service, USDA. Staff Report AGES 9151.
Row, C. (1996) "Effects of Selected Forest Management Options on Carbon Storage." In:
Forests and Global Change. Volume 2: Forest Management Opportunities for Mitigating Carbon
Emissions N. Sampson, and D. Hair (eds.). American Forests, Washington, DC, pp. 59-90.
Row, C., and R.B. Phelps (1991) "Carbon Cycle Impacts of Future Forest Products Utilization
and Recycling Trends." In: Agriculture in a World of Change, Proceedings of Outlook '91, U.S.
Department of Agriculture, Washington, DC, pp. 783-786.
Sampson, R. Neil (2003) "Integrating Land Use, Land Use Change, and Forestry into a
Mandatory National Greenhouse Gas Reduction Program." In: A Climate Policy Framework:
Balancing Policy and Politics. John A. Riggs, (ed.), Aspen Institute, Washington, DC, pp. 63-72.
.
Scheehle, E. and D. Kruger (in press). "Global Anthropogenic Methane and Nitrous Oxide
Emissions." Energy Journal.
Schlamadinger, B. (2003) "Terrestrial Carbon Sequestration: International and U.S.
Initiatives." Presentation at Wildlife Habitat Council, Baltimore. Environmental Synergy
Inc. .
Schneider, U.A. (2000) "Agricultural Sector Analysis on Greenhouse Gas Emission
Mitigation in the U.S." PhD Dissertation, Department of Agricultural Economics, Texas
A&M University, College Station, TX.
Sedjo, R, B. Sohngen, and R. Mendelsohn (2001) "Estimating Carbon Supply Curves for
Global Forests and Other Land Uses." Discussion Paper 01-19, Resources for the Future,
Washington, DC. Available atwww.rff.org/Documents/RFF-DP-01-19.pdf.
Siry, J. (2002) "Intensive Timber Management Practices." In: Southern Forest Resource
Assessment. D.N. Wear and J.G. Greis, eds., Gen. Tech. Rep. SRS 53, Asheville, NC, 635 p.
U.S. Department of Agriculture, Forest Service, Southern Forest Research Station.
-------
REFERENCES
Skog, K.E., and G.A. Nicholson (2000) "Carbon Sequestration in Wood and Paper Products."
In: The Impact of Climate Change in America's Forests: A Technical Document Supporting the 2000
USDA Forest Service RPA Assessment, pp. 79 88. Tech Rep. RMRS GTR 59. Fort Collins, CO.
Smith, W. Brad, John S. Vissage, David R. Darr, and Raymond M. Sheffield (2001) Forest
Resources of the United States, 1997. Gen. Tech. Rep. NC-219. U.S. Department of Agriculture,
Forest Service, North Central Research Station, St. Paul, MN, 191 pp.
Sohngen, Brent, and Robert Mendelsohn (2003) "An Optimal Control Model of Forest Carbon
Sequestration." American Journal of Agricultural Economics 85(2): 448-457.
Stavins, Robert N. (1999) "The Costs of Carbon Sequestration: A Revealed-Preference
Approach." American Economic Review 89(4): 994-1009.
Stavins, Robert N. and Richard Newell (2000) "Climate Change and Forest Sinks:
Factors Affecting the Costs of Carbon Sequestration." Journal of Environmental Economics
and Management 40: 211-235.
Subak, S. (2003) "Replacing Carbon Lost from Forests: An Assessment of Insurance, Reserves,
and Expiring Credits." Climate Policy 3(2): 107-122.
Tanyeri-Abur, A. (1990) "An Agricultural Sector Analysis of the United States Sugar Import
Policy." Ph.D. thesis, Texas A&M University College Station, TX.
Turner, D.P., J.L. Lee, G.J. Koerper, and J.R. Barker. (1993) The Forest Sector Carbon Budget of the
United States: Carbon Pools and Flux under Alternative Policy Options. EPA/600/3-93/093. U.S.
Environmental Protection Agency, Environmental Research Laboratory, Corvallis, OR.
U.S. Department of Agriculture (USDA), Forest Service (2002) Southern Forest Resource
Assessment, A multiagency effort led by the USDA Forest Service's Southern Research Station.
.
U.S. Department of Agriculture (USDA), Forest Service (2003) An Analysis of the Timber
Situation in the United States: 1952 to 2050. Gen. Tech. Rep. PNW-GTR-560. U.S. Department
of Agriculture, Forest Service, Pacific Northwest Research Station, Portland, OR.
U.S. Department of Agriculture (USDA), National Resources Conservation Service (NRCS)
(2000). 1997 National Resources Inventory.
U.S. Department of Energy, Energy Information Administration (EIA). (2004) Annual Energy
Outlook 2004. Report #DOE/EIA-0383 (2004). .
U.S. Department of Labor, Bureau of Labor Statistics (Series) Producer Price Indexes.
.
U.S. Environmental Protection Agency (EPA) (2000) National Water Quality Inventory.
.
U.S. Environmental Protection Agency (EPA). (2003) Inventory of U.S. Greenhouse Gas Emissions
and Sinks, 1990 - 2001. U.S. Environmental Protection Agency, Washington, DC.
U.S. Environmental Protection Agency (EPA). (2005) Inventory of U.S. Greenhouse Gas Emissions
and Sinks: 1990-2002. U.S. Environmental Protection Agency, Washington, DC.
R-8 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
REFERENCES
U.S. Government (2000) "United States Submission on Land-Use, Land-Use Change,
and Forestry." August 1, 2000 .
Uri, N.D. (1997) "State of the Land Update." Natural Resources Conservation Service.
van Kooten, C.G., C.S. Binkley, and G. Delcourt (1995) "Effect of Carbon Taxes and Subsidies
on Optimal Forest Rotation Age and Supply of Carbon Services." American Journal of Agricul-
tural Economics 77: 365-374.
van't Veld, K. and A.J. Plantinga (2005) "Carbon Sequestration or Abatement? The Effects
of Rising Carbon Prices on the Optimal Portfolio of Greenhouse-Gas Mitigation Strategies."
Journal of Environmental Economics and Management 50(1): 59-81.
Vaughan, W.J. (1986) The Water Quality Ladder. Included as Appendix B in Mitchell, R.C.,
and R.T. Carson. 1986. The Use of Contingent Valuation Data for Benefit/Cost Analysis in Water
Pollution Control. CR-810224-02. Prepared for U.S. Environmental Protection Agency, Office
of Policy, Planning, and Evaluation, Washington, DC.
Wassmann, R., R.S. Lantin, H.U. Neue, L.V. Buendia, T.M. Gorton, and Y. Lu (2000). "Charac-
terization of Methane Emissions from Rice Fields in Asia. III. Mitigation Options and Future
Research Needs." Nutrient Cycling in Agroecosystems 58: 23-36.
Wayburn, L., J.F. Franklin, J.C. Gordon, C.S. Binkley, D.J. Mladenoff, and N L. Christensen
(2000) Forest Carbon in the U.S.: Opportunities and Options for Private Lands. Pacific Forest Trust.
Wear, D.N. and B.C. Murray (2004) "Federal Timber Restrictions, Interregional Spillovers,
and the Impact on U.S. Softwood Markets." Journal of Environmental Economics and Manage-
ment 47'(2): 307-330.
West, T.O., G. Marland, A.W. King, WM. Post, A.K. Jain, and K. Andrasko (2004) "Carbon
Management Response Curves: Estimates of Temporal Soil Carbon Dynamics." Environmen-
tal Management 33(4): 507-518.
West, T.O. and WM. Post (2002) "Soil Organic Carbon Sequestration Rates by Tillage and
Crop Rotation: A Global Data Analysis." Soil Science Society of America Journal 66:1930-1946.
West, T.O. and J. Six. (2005) "Considering the Influence of Sequestration Duration and
Carbon Saturation on Estimates of Soil Carbon Capacity." Climatic Change (in press).
Weyant, John P. (2000). An Introduction to the Economics of Climate Change Policy. Prepared for
the Pew Center on Global Climate Change.
Williams, J.R., C.A. Jones, J.R. Kiniry, and O.K. Spanel (1989) "The EPIC Crop Growth
Model." Trans ASAE 32: 497-511.
World Resources Institute and World Business Council on Sustainable Development (WRI/
WBCSD) (2003) The Greenhouse Gas Protocol: Project Quantification Standard. Road Test Draft.
Wu, Junjie (2000) "Slippage Effects of the Conservation Reserve Programs." American Journal
of Agricultural Economics 82: 979-92.
GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE R-9
-------
REFERENCES
Wu, Junjie, David Zilberman, and Bruce A. Babcock (2001) "Environmental and Distribu-
tional Effects of Conservation Targeting Strategies." Journal of Environmental Economics and
Management 41: 333-50.
Zhang, D., J. Buongiomo, and P.J. Ince (1996) "A Recursive Linear Programming Analysis of
the Future of the Pulp and Paper Industry in the United States: Changes in Supplies and
Demands, and the Effects of Recycling." Annals of Operations Research 68:109-139.
R-10 GREENHOUSE GAS MITIGATION POTENTIAL IN U.S. FORESTRY AND AGRICULTURE
-------
'-
«*ij
E
i S*i
w
ft-
r-
-------
Greenhouse Gas Mitigation Potential in U.S. Forestry and Agriculture
United States Environmental Protection Agency
Office of Atmospheric Programs (6207J)
1200 Pennsylvania Ave., NW
Washington, DC 20460
Official Business
Penalty for Private Use
$300
------- |