United States
Environmental Protection
Agency
Envionmental Research
laboratory
Corvallis, OR 97333
EP(V600/3-91/031
April 1991
Research and Development
&EPASequestering Carbon in Soils:
A Workshop to Explore the
Potential for Mitigating Global
Climate Change
LIVING CARBON I DETRITAL CARBON
CO.
CO,
CO
CO,
CK
CO
ABOVEGROUND
BELOWGROUND
EROSION AND LEACHING LOSSES
SIMPLIFIED CONCEPTUAL MODEL OF CARBON POOLS AND FLUXES
IN TERRESTRIAL SYSTEMS
hJ
-------
EP A/600/3-91/031
April 1991
SEQUESTERING CARBON IN SOILS: A WORKSHOP TO EXPLORE
THE POTENTIAL FOR MITIGATING GLOBAL CLIMATE CHANGE
26-28 February, 1990
Corvallis, Oregon, USA
Sponsored by the
US ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Environmental Research Laboratory
Corvallis, Oregon
and the
US ENVIRONMENTAL PROTECTION AGENCY
Office of Policy, Planning, and Evaluation
Office of Policy Analysis
Washington, D.C.
Workshop Coordination and Report by
Mark G. Johnson and Jeffrey S. Kern
With Contributions by
Duane A. Lammers, Jeffrey J. Lee, Leon H. Liegd
Kim Mattson, and Paul Shaffer
USEPA Environmental Research Laboratory
200 S.W. 35th Street
Corvallis, Oregon 97333
April 15,1991
t,4L
-------
DISCLAIMER
The information in this document has been funded wholly or in part by the
U.S. Environmental Protection Agency under contract 68-C8-0006 to
ManTech Environmental Technology International, Inc. It has been subjected
to agency's peer and administrative review, and it has been approved for
publication as an EPA document. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
COVER GRAPHIC
The cover graphic depicting a simplified conceptual model of carbon pools
and fluxes in terrestrial systems is the product of discussions between M.G.
Johnson and K.G. Mattson.
ACKNOWLEDGMENTS
We acknowledge the participation and contributions of everyone who took
part in the workshop on soils and climate change. We are grateful to Ken
Andrasko and Kate Heaton of the EPA's Office of Policy Planning and
Evaluation (OPPE) in Washington, D.C., for their curiosity about soils and for
OPPE's financial sponsorship of the workshop. We are particularly indebted
to Michael Beare, Sandra Brown, Stan Buol, Dale Cole, Kermit Cromack,
John Duxbury, Richard (Skee) Houghton, Dale Johnson, Rattan Lai, Ariel
Lugo, Bill Parton, Mac Post, Paul Rasmussen, Bill Schlesinger, and Phil
Sollins for their thought-provoking presentations, contributions, or reviews.
We also thank Chris Andersen, Deborah Coffey, Kate Dwire, Ann Hairston,
Duane Lammers, Leon Liegel, Jeff Lee, Kim Mattson, Paul Shaffer, Bill
Ferrell, Hermann Gucinski, Jack Winjum, and Bob Dixon for their reviews
and contributions to this document. Finally, we are thankful for the
encouragement and patience of Bruce McVeety and Dave Tingey who have
helped us stay on track and complete this report.
M.GJ and J.S.K, Corvallis, OR
-------
APPENDIX B: PRESENTATION SUMMARIES 25
Stan Buol 25
William Schlesinger 26
Richard Houghton 28
Ariel Lugo 29
Phil Sollins 31
John Duxbury 32
Michael Be are 34
W. Mac Post 35
Rattan Lai 36
William Parton 38
Paul E. Rassmussen 40
Kermit Cromaek 41
Dale Johnson 42
APPENDIX C; SUBMITTED PAPERS 43
Beare, M.H. and P.P. Hendrix; Possible Mechanisms for the Accumulation and Loss of Soil
Organic Carbon in Agroeeosystems on the Southern Piedmont 43
Buol, S.W.: Pedogenesis of Carbon in Soils 52
Lai, R., Managing Soil Carbon in Tropical Agro-Ecosystems 56
Lugo, A.E., and S. Brown, Management of Tropical Forest Lands for Maximum Soil Carbon
Storage 75
APPENDIX D: LETTER FROM DAVID JENKINSON . . 78
APPENDIX E: LETTER FROM HANS JENNY 82
APPENDIX F: WORKSHOP PARTICIPANTS 83
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY i
1. INTRODUCTION 1
2. WORKSHOP GOALS AND ORGANIZATION 1
Workshop Goals 1
Workshop Report Objectives and Organization . 2
1 THE GLOBAL CARBON CYCLE 2
Global Carbon Pools and Fluxes . 3
Mitigation 3
Unknowns 3
4. SOIL CARBON POOLS 3
Global Distribution of Soil Carbon 4
Genesis of Soil Carbon 4
Soil Carbon Accumulation Rates 5
Loss of Soil Carbon 6
Conceptualizing and Quantifying Soil Carbon Pools . 6
Forecasting Soil Carbon Dynamics 7
Unknowns 7
5. MANAGING SOIL CARBON 8
Carbon Sequestration in Soils 9
Carbon Carrying Capacity 9
Decomposition 9
Temperature Effects 9
Soil Fertility 9
Soil Aggregation 10
Agricultural Systems: Temperate and Tropical 10
Forested Systems: Temperate and Tropical 10
Unknowns 11
6. STRATEGIES FOR MANAGING SOIL CARBON AT THE GLOBAL SCALE 11
MAINTAINING THE GLOBAL POOL OF SOIL CARBON 12
Management Practices to Maintain Soil Carbon Levels 12
RESTORING SOIL CARBON IN CARBON-DEPLETED SOILS 14
Management Practices to Restore Carbon in Carbon-Depleted Soils . 14
ENLARGING THE SIZE OF THE GLOBAL SOIL CARBON RESERVOIR 16
Management Practices for Enlarging the Global Soil Carbon Reservoir 16
7. WORKSHOP SUMMARY AND CONCLUSIONS 17
8. RESEARCH RECOMMENDATIONS 18
LITERATURE CITED 18
APPENDIX A: WORKSHOP AGENDA 1 21
/¦a/
-------
EXECUTIVE SUMMARY
SEQUESTERING CARBON IN SOILS: A WORKSHOP TO
EXPLORE THE POTENTIAL FOR MITIGATING GLOBAL CLIMATE CHANGE
INTRODUCTION
Soils are an important component of the global
carbon cycle and a major reservoir of carbon.
Oxidation of soil carbon through agriculture,
deforestation, and changing land use practices
contributes to the buildup of atmospheric carbon
dioxide and methane. Increases in atmospheric
concentrations of greenhouse gases, such as carbon
dioxide and methane, are causing additional solar
energy be retained in the earth's atmosphere
leading to global warming that may alter global
climatic patterns (Mitchell, 1989). Strategies to
mitigate global warming include reducing
greenhouse gas emissions (U.S. Congress, 1991) and
sequestration of atmospheric carbon in terrestrial
vegetation (Houghton, 1990; Grainger, 1990).
Because soils can function as either a source of
atmospheric carbon or a sink, and because they are
the growth medium of terrestrial plants, they may
have an important role in mitigating global
warming. Because the linkage between soils and
climatic change is not fully known a workshop was
convened to address the question "Can soils be used
to store sufficient carbon to aid in mitigating global
climate change?".
Figure 1
SCIENTIFIC SETTING
Recent scientific and public concerns have focused
on the potential for global warming and global
climatic change. The principal reason for these
concerns is the documented increase in greenhouse
gases in the atmosphere. These gases absorb solar
energy and translate it into thermal energy, resulting
in some degree of atmospheric warming.
Greenhouse gases include water, carbon dioxide,
methane, nitrous oxide, and chlorofluorocarbons.
Carbon dioxide, methane, and nitrous oxide are
naturally occurring greenhouse gases, yet their
increased concentrations are linked with human
activities [Figure 1], Of particular interest is carbon
dioxide because its increase is correlated with fossil
fuel combustion and biomass burning associated
with large-scale deforestation.
GLOBAL RESERVOIRS OF CARBON
(Petagrams of Carbon)
Vegetation 550
Soils 1,500
Atmosphere 750
Oceans 38,000
Recoverable Fossil Fuels 4,000
Carbon circulates between three very large
reservoirs (oceans, atmosphere, and terrestrial
systems) and can be found in a plenitude of
compounds in each reservoir. These reservoirs, or
pools, exchange large amounts of carbon annually.
A fourth reservoir, the geolojpcal reservoir,
contains fossil and mineral carbon including
carbonates, and consists primarily of inactive or
non-circulating carbon. Perturbations, disturbances,
or additions of carbon (e.g. fossil fuel combustion)
to any of the reservoirs will have a concomitant
effect on the others because of the dynamic linkage
of the reservoirs. Global warming is also expected
ANTHROPOGENIC CONTRIBUTION TO
RADIATIVE FORCING - 1980 TO 1990
Methane - 16%
Nitrous Oxide - 6%^
Other CFGs - 716 A |
\ CFCa f! and 12 - 17*
Carbon Dioxide
- 55%
Houghton tt *1. 1990
-------
to have am effect on the balancing of the global
carbon cycle. One projected effect is the shifting of
global vegetation and the amount of carbon stored
therein.
OsfBr»3£a.Udft
Atmosphere; 750 +3/year
. ftiv&rs
Larva Biota
Soli ar* Dei-H
^500
FoasH
do
Surface Oesflfc *>00 * i/year
Intermediate and De&p water*
38,000
Figure 2. Current estimates of global caibon pools and fluxes in
petagrams of cartx>n including annual increases due to human
activities, (source: Houghton ct aL, 1990)
Globally, there are approximately 41,000 Pg (Pg =
petagrams = 1015 grams) of active, or circulating,
carbon [Figure 2]. Of this, the oceanic reservoir
contains 38,000 Pg; the atmosphere 750 Pg; and
terrestrial ecosystems about 2100 Pg of carbon. Of
the terrestrial carbon, living plants account for
about 550
Pg
of carbon and soils approximately
1500 Pg. Soils are therefore the largest, non-fossil,
terrestrial reservoir of carbon. Annually, the
current net loss of carbon from plants and soils is
estimated to be 1 Pg. The cumulative loss over the
last 100 or 200 years has been considerable, but
there is a great deal of uncertainty in the actual
amount of carbon lost due to the lack of reliable
data and the heterogeneity of the earth's soils and
vegetation.
WORKSHOP ON SOIL CARBON
The US Environmental Protection Agency (EPA)
sponsored this workshop to evaluate the potential of
soils to sequester and store carbon. The workshop
was attended by 40 scientists, including
internationally recognized soil and carbon cycling
experts. The workshop was held in Corvallis,
Oregon and consisted of two days of informal
presentations and discussions, preceded by a one-
day field trip to observe agricultural and forestry
practices that affect above- and belowground carbon
storage.
The workshop had three specific goals:
(1) to provide the EPA with an informed scientific
opinion as to whether or not carbon can be
sequestered in soils on a global scale to the
extent that it can be used to reduce the rate of
atmospheric carbon dioxide increases;
(2) to identify the major unknowns regarding this
opinion;
(3) to recommend research that the EPA should
pursue to eliminate knowledge gaps and
uncertainties in this area.
To achieve these goals, the workshop was organized
around three key areas: (1) the global carbon cycle,
(2) soil carbon pools, and (3) managing soil carbon.
Scientists were invited to speak in each of these
areas, emphasizing what is known, identifying
uncertainties and limitations, and making research
recommendations. Agricultural and forestry
practices were accentuated because of the potential
for using them or adapting them for managing soil
carbon.
WORKSHOP FINDINGS
In setting the stage for deliberations on the
mitigating potential of soils, the speakers conveyed
important background information. This
information is summarized here.
• GLOBAL CARBON CYCLE: At steady state,
the flow of carbon between the global carbon
pools is in equilibrium. The amount of carbon
fixed annually by terrestrial plants through
photosynthesis ranges from 100 -120 Pg. Plant
respiration releases approximately 40 - 60 Pg of
carbon annually, and decomposition of organic
residues, including soil carbon, releases
approximately 50 - 60 Pg. At steady state the
amount of carbon oxidized by these two
processes would balance that fixed by
photosynthesis. Through agricultural practices
and land use changes, including deforestation,
the oxidation of plant and soil carbon may be
exceeding the amount being fixed by
photosynthesis, and therefore contributing to
the net 3 Pg annual increase in atmospheric
carbon dioxide.
One of the concerns and unknowns associated
with global warming and the carbon cycle is the
effect of warming on the distribution of carbon
-------
in the various reservoirs. Will increased
temperatures cause a massive shift in carbon
now sequestered in soils to the atmosphere
thereby increasing global warming? It was
reported that if soil temperatures increase 3°C,
approximately 10% of the carbon in temperate
and agricultural soils (approximately 50 Pg)
could be released into the atmosphere. A 10%
loss of all the carbon currently held in soils
would result in ISO Pg of carbon being injected
into the atmosphere. Either scenario would
exacerbate the amplitude and extent of global
warming. If half of a percent were lost per
year, it would only take 20 years.
SOIL CARBON POOLS: The accumulation
and distribution of soil carbon depend on a
variety of biotic and abiotic factors including;
(1) soil chemical and physical characteristics;
(2) precipitation; (3) above- and belowground
biology; (4) temperature; (5) solar radiation; (6)
atmospheric chemistry and processes; (7)
landscape characteristics; (8) site history; and
(9) time. Land use practices affect these
factors and thus affect soil carbon.
LIVING CARBON I DETftJTAL O.R60N
CO A CO n
CO ,
CO ,
ABOVEGBOUND >¦
LITTER
ABOVE ROUND
BELOWGROUND
BELOWGROUND
CO,
UNPRtstgeno
SOIL CABBON
PROTECTED
SOIL CARBON
EROSION AND LEACHiNG LOSSES
Figure 3. Simplified conceptual model of carbon pools and fluxes
III terrestrial systems.
Carbon fixation via photosynthesis is the
ultimate source of soil carbon and provides the
energy that drives soil biological processes
[Figure 3]. The non-mineral carbon in soils
(organic carbon) can be associated with either
living organisms or their residues. Living soil
organisms include plant roots, macroorganisms
or fauna, and microorganisms. Combined these
comprise less than four percent of total soil
carbon. The remaining 96 - 98% of soil carbon
is detrital and consists of about 20%
macroorganic matter and 80% humified
(partially decomposed or altered) material.
DISTRIBUTION OF CARBON IN WORLD SOILS
(Petagrams of Carbon)
Total • 1396 Petaarams
Boreal ForesSe
161,e
Savanna
1286
Thorn Steppa
28.6
Cool Steppe
Desert
84
Tundra
wetlands
202,4
Temperate forests
104,3
t Tropical Forests
184.6
Cultivated Land
18?,5
Post at 81., 1902
Figure 4.
The carbon content of soils can be quite high,
as in the case of wet peat or muck soils
(Histosols) with carbon contents as high as 72.3
kg carbon m"2 [Figure 4]. It can also be quite
low as in dry desert soils (Aridisols) with values
as low as 1.4 kg carbon m"2. Soil carbon
oxidation rates are subject to soil temperature,
oxygen supply, and the nature of the organic
material. The difference is due primarily to soil
moisture and temperature. Soil carbon
contents tend to increase curvilinearly with
increasing rainfall at a constant mean annual air
temperature (Jenny, 1980). However, at a
constant moisture content soil carbon content
tends to decrease curvilinearly with constant
temperature. Soils with mean summer and
winter temperatures (measured at a depth of 50
cm) that differ less than 5°C tend to have more
soil carbon than soils with more than 5°C
difference but the same mean annual
temperatures. Soils saturated with water for
long periods of time have low oxygen contents
and therefore low decomposition rates and
carbon tends to accumulate. Other factors such
as clay content and mineralogy also affect soil
carbon contents.
An important characteristic of soil carbon is its
retention time or turnover time. Retention
times are a measure of stability of organic
matter under existing conditions. Soil carbon
retention times range from as short as a few
years or as long as thousands of years, with the
longer times being associated with cold and wet
climates. Carbon retention times are also a
iii
-------
function of depth within the soil. Organic
material on, or near, the soil surface is
susceptible to decomposition. Materials
contained within the soil profile tend to be
more protected and less susceptible to
decomposition. For example, the carbon in
forested soils is derived from leaf litter
deposited on the soil surface and from fine root
turnover near the soil surface. The primary
source of carbon in grassland soils is root
mortality which is incorporated within the soil
to depths of a meter or more. For this reason,
within a given climate regime, a grassland soil
will generally have more carbon (longer
retention times) than a forested soil even
though the forested system will have greater net
primary production and aboveground biomass
because of depth of organic matter
incorporation.
SOIL CONDITIONS AND MANAGEMENT
PRACTICES THAT PROMOTE CARBON
ACCUMULATION IN SOILS
Soil Change ' Practice , Applications
Cooler soil . Mulch, shade"' • All soils
Wetter soil ¦ Irrigation Dry regions
Increase Fertility Fertilizer " - Most soils
Raise subsoil pH .Deep liming Acid subsoils
Reduced aeration Limited tillage All soils
(source S.W. Buol and UNEP) ,
• MANAGING SOIL CARBON: In general, soils
have not been managed to retain or conserve
carbon. Although economics, as measured by
crop yields and board feet of lumber, has been
the common metric driving agricultural and
forest system management, soils are an
important component in the production
equation. In addition to being an important
carbon reservoir, organic carbon in soils is
important for maintaining the productivity of
soils. Organic carbon in soil contributes to soil
productivity by: promoting soil aggregation,
absorbing and holding water, serving as a
natural reservoir of plant nutrients, minimizing
wind and water erosion, and providing exchange
sites for plant nutrients.
Soil fertility is essential for primary production
and is therefore key to sequestering carbon in
soils. Soils that have lost carbon or that are
naturally low in fertility have the potential to
store carbon by improving their fertility status.
To sequester carbon in soils, fertility limitations
need to be identified and eliminated. Judicious
fertilizer use is recommended, however,
because even though more carbon may be
sequestered in soils there may be negative
effects too and there are carbon costs
associated with fertilizer production and
application. Use of nitrogen fertilization may
promote the emission of nitrous oxide from
soil. Per molecule, the radiative forcing effect
of N20 is about 200 times greater than C02
(U.S. Congress, 1991).
The amount of organic carbon b soils is a
function of the quantity and quality of carbon
inputs and subsequent losses through
decomposition and erosion. Soil carbon is also
a function of soil chemical and physical
properties. Increasing the inputs of carbon to
soils can increase the amount of carbon
sequestered and immobilized in soils.
Minimizing soil mixing and warming will slow
decomposition. Managing agricultural soils to
conserve and store carbon requires the use of
conservation tillage practices to minimize soil
disturbances, and the, incorporation of crop
residues into the soil to increase carbon inputs
to the soil. Soil temperature is positively
related to decomposition. The use of mulches
or cover crops reduce soil temperatures,
thereby reducing soil carbon losses.
Although forest management is markedly
different than agricultural management, the
principles of minimizing soil disturbance and
taking steps to lower soil temperatures are still
applicable for conserving carbon in forested
soils. The removal of forest cover results in
increased oxidation of soil carbon through
increased soil temperatures, particularly in large
exposed clear-cuts. Soil disturbances associated
with forest harvesting also contribute to the net
loss of soil carbon. Implementing management
practices that minimize soil disturbance during
forest harvesting and providing soil mulches
following harvesting will conserve soil carbon in
forested soils.
As soils near their carbon carrying capacity, or
equilibrium carbon content, the carbon
accumulation rate becomes very low
iv
-------
(Schlesinger, 1991). Similarly, soils forming in
newly exposed geologic material accumulate
carbon slowly because they lack the functional
organization to support high levels of primary
production, but with time the rate will increase.
The soils having the greatest carbon
accumulation rates are those that are not at
their carbon carrying capacity which includes
young soils and soils that have been somewhat
carbon depleted due to management.
Currently there is about 1.5 billion hectares of
cropland (World Resources Institute, 1990).
Conservation tillage could be implemented on
many of these soils, thereby reducing soil
carbon losses and converting these soils from
sources of atmospheric carbon to sinks.
Additionally, with proper management carbon
could be restored to many other carbon-
depleted soils. In the tropics there is an
estimated 865 million hectares of deforested
and abandoned land that could be is potentially
available for afforestation (Houghton, 1990). If
reforested, these systems would withdraw
approximately 1.5 Pg carbon a year from the
atmosphere sequestering it in soils and
vegetation over the next century. Similar
opportunities exist in other regions of the world
as well.
STRATEGIES FOR MANAGING SOIL CARBON
AT THE GLOBAL-SCALE
Workshop participants recognized that some soils
could be used to store additional carbon, but based
upon currently available data, could not definitively
conclude whether or not the storage of new carbon
in soils would be sufficient to offset atmospheric
increases in carbon or to mitigate global warming.
They emphasized the importance of managing soils
to prevent soils from continuing to be a source of
greenhouse gases, thereby reducing the amplitude
and extent of global warming.
Workshop participants identified three general soil
carbon management strategies to optimize carbon
sequestration and storage: (1) manage soils to
maintain the size and integrity of the global pool of
soil carbon, (2) manage soils to restore carbon that
has been lost, and (3) manage soil to increase the
amount of carbon sequestered in the global soil
carbon reservoir. Each of these described below,
followed by descriptions of specific management
practices that the workshop participants determined
would be important for achieving the soil
management goals.
SOIL CARBON MANAGEMENT
'"''STRATEGIES,,;.
• Manage to MAINTAIN8 the global soil
carbon pool
• Manage to RESTORE carbon to carbon-
depleted soils
• Manage to ENLARGE the global soil
carbon pool :
MAINTAINING THE GLOBAL POOL OF SOIL
CARBON; Because soils are the largest terrestrial
pool of active, or cycling, carbon and because a
large portion of this carbon is potentially available
to the atmosphere through human activities and
conventional soil management practices, workshop
participants recommended that management
practices be implemented to preserve the size and
integrity of the pool. Managing soils to maintain
their current carbon levels recognizes that soil
carbon is labile and is lost to the atmosphere
relatively easily when the soil is mixed and stirred as
in conventional agricultural systems, On the other
hand it takes a substantially longer even with
sustained carbon inputs to accumulate significant
amounts carbon in soils. Workshop participants
concluded that it is best to protect and preserve sod
carbon before it is lost, than it is to allow it to be
lost and then try to restore it. They also recognized
that a significant loss of soil carbon to the
atmosphere could exacerbate global warming.
HIGH PRIORITY MANAGEMENT
PRACTICES FOR MAINTAINING
GLOBAL SOIL CARBON POOLS
• Maintain Soil Fertility
• Concentrate Tropical Agriculture
• Preserve Natural Wetlands
• Increase Efficiency of Forest Product
Use
-------
Management Practices to Maintain Soil Carbon
Levels
The management practices described below were
identified by the workshop participants as those that
could be used to maintain the amount of carbon in
soils and were given the highest priority for
implementation.
• Maintain Soil Fertility: Sustained inputs of
new carbon to soils from primary production is
essential for maintaining current levels of soil
carbon. Interruption of these carbon inputs
results in loss of soil carbon by the
readjustment of the soil system to a lower
equilibrium level of soil carbon. Maintaining
soil fertility helps sustain primary production,
which in turn helps to sustain the input of new
carbon belowground and the sequestration of
carbon in soils.
• Concentrate Tropical Agriculture:
Concentrating or intensifying tropical
agriculture is aimed at preventing tropical
deforestation and the soil carbon loss associated
with it, and thereby maintaining the existing
stocks of soil carbon. Concentrating tropical
agriculture is the practice of directing resources
and management for use only on the best
agricultural lands. The net result is that less
land will be used to produce the same, if not
more, volume of agricultural products. In
achieving this, the need for slash-and-burn
agriculture (deforestation) is diminished.
• Preserve Natural Wetlands: Inundated soils in
wetlands contain a great deal of carbon that has
a very long retention time if undrained. The
practice of draining these systems, primarily for
agrarian purposes, results in the rapid oxidation
and loss of carbon to the atmosphere.
Preserving wetlands insures the carbon storage
function of these lands.
• Increase Efficiency of Forest Product Use:
Increasing the efficiency of forest harvesting (by
reducing waste) and increasing the life-span of
forest products (including recycling) is aimed at
reducing the demand for forest harvesting.
Reducing forest harvesting, or increasing the
interval between harvests, will help to maintain
the large above- and belowground carbon
stocks in the worlds forests.
RESTORING SOIL CARBON IN CARBON-
DEPLETED SOILS: An opportunity exists to
restore carbon in soils that are depleted in carbon
due to mismanagement. Worldwide, there may be
millions of hectares of carbon-depleted forest and
agricultural land that could store additional carbon.
Management practices that bring these once
productive lands back into increased production will
lead to increased carbon sequestration and storage
both above- and belowground. Soils that are the
most carbon depleted will require more inputs to
restore them to their full carbon carrying capacity.
The long-term objective of this strategy is to restore
the soil carbon levels to what they were originally.
HIGH PRIORITY MANAGEMENT
PRACTICES FOR RESTORING
CARBON IN CARBON-DEPLETED
SOILS
• Reforestation
• Improve Soil Fertility
• Use Municipal, Animal, Industrial and
Food Processing Wastes as a Source
of Low-Cost Fertilizer
Management Practices to Restore Carbon in
Carbon-Depleted Soils
The management practices described below were
identified as those that would be useful for restoring
soil carbon to previous levels and were given the
highest priority for implementation to restore
carbon in carbon-depleted soils.
• Reforestation: Large-scale reforestation has
the potential to sequester and store large
amounts of new carbon both above- and
belowground, particularly on carbon-depleted
soils. Some deforested areas will regenerate
naturally while others will only require new tree
planting. Others will require intensive
management and large inputs of resources.
• Improve Soil Fertility: Soils that are carbon
depleted may also lack in soil nutrients, which
may limit primary production, Improving the
fertility of these soils to support vegetation is
key to restoring carbon in these soils.
-------
• Use Municipal, Animal, Industrial and Food
Processing Wastes: Restoring carbon in
carbon-depleted soils may be limited by the
lack of plant nutrients needed for primary
production or by the cost of fertilizer to
replenish nutrients. Municipal, animal,
industrial, and/or food processing wastes may
be ideal sources of low-cost forest fertilizers.
Using them in forests or in agriculture,
particularly on marginal lands, can provide
water and essential nutrients to vegetation,
thereby promoting a higher level of primary
production and above- and belowground carbon
sequestration.
ENLARGING THE SIZE OF THE GLOBAL SOIL
CARBON RESERVOIR: The objective of this
strategy is to manage forested and agricultural
systems in ways that increase their productivity and
their allocation and storage of carbon belowground.
Most of the opportunities to enlarge this pool may
be in agriculture because agricultural systems are
generally more intensely managed than forested
systems. Workshop participants surmised that the
marginal return, measured in terms of stored
carbon, would probably be greater by implementing
management practices aimed at "maintaining" and
"restoring" storing soil carbon than by attempting to
enlarge the global pool of soil carbon.
HIGH PRIORITY MANAGEMENT
PRACTICES FOR ENLARGING THE
GLOBAL POOL OF SOIL CARBON
• Conservation Tillage
• Improve Soil Fertility
• Concentrate Tropical Agriculture
• Minimize Dryland Fallowing
Management Practices for Enlarging the Global
Soil Carbon Reservoir
The management practices described below were
given the highest priority for implementation to
enlarge the global reservoir of soil carbon.
• Conservation Tillage: In the context of
enlarging global soil carbon stocks, widespread
implementation of conservation tillage practices
could lead to additional sequestration of carbon
in soils. Implementing conservation tillage
practices implies that the use of conventional
tillage practices will decline, thus reducing the
oxidative loss of soil carbon.
• Improve Soil Fertility: Lack of nutrients often
limits primary production. Eliminating or
reducing nutrient limitations will improve
primary production leading to greater
sequestration of carbon in soil.
• Concentrate Tropical Agriculture: By
concentrating tropical agriculture, lands
removed from shifting agriculture can be
reforested or revegetated, leading to increased
soil carbon stocks.
• Minimize Dryland Fallowing: Fallowing is the
practice of leaving semi-arid agricultural lands
bare in alternate years to accumulate sufficient
soil moisture to grow a crop every other year.
Dryland fallowing is a widely used agricultural
practice. There is gathering evidence that this
practice promotes the loss of soil organic
carbon because the soil is usually kept bare by
mechanical means. This stirring and mixing of
the soil combined with increased soil
temperatures, due to lack of cover, results in
rapid and thorough oxidation of organic carbon.
Use of cover crops or crop rotations may
achieve the same soil water objectives and
eliminate the need to fallow.
CONCLUSIONS
Two primary conclusions can be drawn from the
workshop. First, that steps should be taken to
protect and preserve the size and integrity of the
global reservoir of soil carbon because continued
losses of soil carbon to the atmosphere could
exacerbate global warming and climatic change.
Second, that steps should be taken to manage soils
and ecosystems to store additional carbon. The
latter is to be accomplished predominately by
increasing net primary production. The workshop
identified the major uncertainties related to carbon
sequestration in soils and developed specific
strategies for addressing the uncertainties and for
managing soils to store carbon.
Major conclusions of the workshop participants;
• Soils are an important component of the global
carbon cycle, containing a large pool of active,
vii
-------
cycling carbon. As such, they are a very large
potential source of atmospheric carbon, but
they also represent a large potential sink for
carbon, if managed properly.
• Uncertainties exist in the success of widespread
implementation of soil management practices
because of the potential for the occurrence of
concomitant negative effects. At some locations
the associated negative effects may outweigh
the positive benefits. For instance, under
certain circumstances some of these practices
could lead to the emission of gases (e.g., nitrous
oxide) that have a greater radiative forcing than
carbon dioxide. In terms of global warming,
this scenario would be counterproductive. The
implementation of any new or altered
management practices should be considered on
a site-by-site or region-by-region basis prior to
implementation, and should be evaluated in
terms of the effect on carbon pools and fluxes.
• Three strategies for managing soil carbon are
proposed; (1) managing soils to maintain
current levels of soil carbon, (2) managing
carbon depleted soils to restore carbon to
former levels, and (3) managing soils to
enhance the size of current soil carbon pools.
For each of these a variety of opportunities
exist for capturing atmospheric carbon, via
photosynthesis, and storing it in soils and
aboveground biomass.
• In addition to storing carbon assimilated from
the atmosphere, managing soils to conserve
carbon will have other benefits. These include:
(1) increased soil water holding capacity, (2)
increased nutrient availability, (3) improved soil
physical properties, and (4) decreased soil
erosion by wind and water. Together these
should lead to (5) sustainable food and fiber
production.
The consensus on the central workshop question
"Can soils be used to store sufficient carbon to aid
in mitigating global climate change?" is that we need
more reliable information to provide a definite
answer.
RESEARCH RECOMMENDATIONS
Throughout this workshop, the lack of sufficient,
reliable, quantitative data and numerous areas of
uncertainty related to soil carbon were identified.
Research and data gathering in these areas will
improve our quantification of global soil carbon and
related processes. Here we list those topics thought
to be critical for a more complete analysis of soils
and global climatic change. This list, although not
exhaustive, provides guidance for research in areas
that could provide valuable information or tools for
evaluating the role of soils in the global carbon
cycle with a focus towards the potential of soils to
sequester and store additional carbon from the
atmosphere.
• Define soil carbon pools based upon lability and
soil processes and determine the factors that
control the partitioning of carbon into the
respective pools.
• Identify and characterize the specific fractions
of soil carbon that are manageable and
practices that are effective for managing the
various fraction.
• Improve estimates of global soil carbon by
conducting large-scale statistically designed soil
surveys coupled with intensive soil sampling and
physical and chemical analysis.
• Quantify above- and belowground carbon pools
and fluxes in specific ecosystems under steady
state conditions for the purpose of developing
general principles of ecosystem carbon
dynamics from specific examples.
• Characterize and quantify the factors that
control soil carbon fluxes.
• Develop soil carbon methods for
characterization and quantifying the true size
and lability of soil carbon pools.
• Characterize and quantify the role of abiotic
soil factors in stabilizing soil carbon,
• Quantify the effects of land use and
management, including agricultural and
silvacultural, on soil carbon.
• Conduct experiments to characterize the effects
of altered climatic conditions on terrestrial
plant carbon fixation and allocation, focusing on
the quantity and quality of carbon in detritus
and in belowground allocation.
• Develop simulation models that accurately
project how carbon fluxes (and thus pools and
feedbacks to the atmosphere) will shift in
specific ecosystems under a series of altered
climate scenarios.
-------
• Quantify the economics of implementing soil
management practices that sequester and store
carbon.
EPILOGUE
It was the general consensus of the workshop
participants that because global warming and
climatic change are international problems, a major
global-scale effort is needed to grasp the full scope
of these problems and to conduct the research and
planning needed to develop global-scale solutions.
As research is conducted on global climatic change
and mitigation plans are formulated, it is important
that the role of soils and other components of
terrestrial ecosystems be considered because of the
dynamic linkages between the components and
climate.
The storage of carbon in soils is a very complex
phenomenon. Although it is not fully characterized
or understood, steps can be taken to use soils as a
reservoir of carbon. The role of soils in the carbon
cycle must be more fully understood to develop
strategies to mitigate increases in atmospheric
carbon dioxide, including strategies to manage the
biosphere. Likewise, the effects of specific
management practices on soil carbon cycling and
storage must be more fully understood. The gaps
in our knowledge of soils and ecosystems, and their
response to climatic change, necessitate additional
research before quantifiable projections of the role
of soils in global climatic change can be made.
This workshop identified the major uncertainties in
our understanding of the role of soils in global
climate change and identified steps that can be
taken to manage soils to optimize their carbon
storage potential and sustain their productivity.
January 1990, Conference Proceedings, U.S. Environmental
Protection Agency, Washington, D.C.
Jenny, H. 1980. The Soil Resource: Origin and Behavior.
Springer-Veriag, New York.
Mitchell, J.F.B. 1989. The "Greenhouse" Effect and Climate
Change. Reviews of Geophysics 27:115-139.
U.S. Congress. 1991. Changing by degrees: Steps to reduce
greenhouse gases. Office of Technology Assessment, OTA-O-432
Washington, D.C.
World Resources Institute. 1990. In Hammond, A.L* et al. (ed.)
World resources 1990-91, A report by the World Resources
Institute, Oxford University Press, Oxford, England
CITED REFERENCES
Grainger, A. 1990. Modeling the impact of alternative
afforestation strategies to reduce cartoon dioxide emissions, p,
93-1M. In Tropical Forestry Response Options to Global
Climate Change, Sao Paulo, January 1990, Conference
Proceedings, U.S. Environmental Protection Agency,
Washington, D.C.
Houghton, J.T., GJ. Jenkins, and JJ. Ephraums (eds). 1990.
Policymakers Summary, Jn Climate Change: The IPCC
Scientific Assessment. Cambridge University Press, Cambridge,
England,
Houghton, RA. 1990. Projections of Future deforestation and
reforestation in the tropics, p. 87-92. Jn Tropical Forestry
Response Options to Globai Climate Change, Sao Paulo,
-------
SEQUESTERING CARBON IN SOILS: A WORKSHOP TO
EXPLORE THE POTENTIAL FOR MITIGATING GLOBAL CLIMATE CHANGE
1. INTRODUCTION
Long-term atmospheric chemistry data indicate that
human activities have increased the concentration of
a number of greenhouse gases due to fossil fuel
combustion and land use changes, that may be
enhancing the absorption the earth's long-wave
black-body radiation (Houghton et al., 1990), This
increased retention of solar energy through the
"greenhouse" effect, may lead to global warming and
changes in global climates through radiative forcing.
Long-term earth surface temperature data indicates
that the surface of the earth may have warmed as
much as 0.3°C - 0.6°C in the last century (Houghton
et al., 1990). This temperature increase is further
substantiated by evidence of retreating glaciers and
sea level rise. Because the extent, duration, and
amplitude of global warming is largely unknown, the
long-term effect on global climate is also unknown.
ANTHROPOGENIC CONTRIBUTION TO
RADIATIVE FORCING - 1980 TO 1990
Meshene - 16%
NIkous Oxide - 6%
Carbon D!a*)de - 55%
Heufihton it »t. 1990
Figure 1
The global carbon cycle has a central role in global
climatic change because two of the important
greenhouse gases, carbon dioxide (COj) and
methane (CH4), are part of the actively cycling
carbon. Deforestation and other land use practices
release large amounts of non-fossil fuel carbon into
the atmosphere, thus adding to the increase in
greenhouse gases. Soils are an important
component of the carbon cycle, providing a growth
medium for plants and storing a large quantity of
carbon belowground.
In response to concern over the possibility of global
climatic change, the U.S. Environmental Protection
Agency (EPA) and other federal agencies are
gathering information and conducting research to
determine the likelihood and potential extent of
climate change, the effects of climatic change, and
strategies to mitigate climatic change. In this report
we present the findings and recommendations of a
workshop that was convened to examine the role of
soils in the global carbon cycle and to consider the
potential of soils to mitigate global climate change
by the sequestration of additional atmospheric
carbon in soils.
2. WORKSHOP GOALS AND ORGANIZATION
The workshop on soils and climate change was co-
sponsored by the Climate Change Division of EPA's
Office of Policy Analysis, Washington, D.C., and the
EPA's Environmental Research Laboratory in
Corvallis, Oregon. The workshop was held
February 27 and 28,1990, on the campus of Oregon
State University. More than 40 scientists from
universities, and federal and private research
organizations participated in the workshop,
including internationally recognized soil and carbon
cycling experts. The participants and their
affiliations are listed in Appendix F.
Workshop Goals
The workshop was organized around a central
question, "Can soils be used to store sufficient
additional carbon to aid in mitigating global climate
change?*. To answer this question three workshop
goals were stated at the outset:
(1) to provide the EPA with an informed
scientific opinion as to whether or not
carbon can be sequestered in soils on a
global scale to the extent that soils can be
used to reduce the rate of atmospheric
carbon dioxide increases.
(2) to identify the major unknowns regarding
the sequestration of additional atmospheric
carbon in soils.
(3) to recommend research that should be
pursued to eliminate knowledge gaps and
uncertainties in this area.
To achieve these goals, the workshop was organized
around three key areas:
l
-------
• the global carbon cycle
• soil carbon stocks
OatorMtatlan
Atrncsphara: 750 +3/yeaf
1D2
50
90
50
Surface Ocean: 100G * Vyeat
I ai°"'3 z£z %
Soil and Cel'ltjf
1500
Inter and D«*D W»t«6|
38,000 »2/yw !
Sedimentation 10.1
• managing soil carbon
Scientists were invited to speak in each of these
areas and asked to highlight what is known, identify
uncertainties and limitations, and make research
recommendations. Agricultural and forestry
practices were accentuated at the workshop because
of the potential for using them or adapting them for
managing soil carbon. Each of the presentations
was followed by a period for questions and
discussion. To facilitate additional scientific
discussion, participants were organized into three
small discussion groups and were provided with a
set of questions to guide the discussions. One
group focused their discussions on forested soils,
one on soils in temperate agriculture, and one on
soils in tropical agriculture. These groups met
several times during the workshop. The workshop,
agenda is reproduced in Appendix A.
Workshop Report Objectives and Organization
The purpose of this report is to provide
documentation and a synthesis of the information
presented in the workshop. This report is written to
be used as a source of information for policy
makers and scientists. It includes a list of specific
research topics that, if pursued, would reduce
knowledge gaps and uncertainties in current soil
carbon data and management practices.
This report is organized around the three focus
areas: the global carbon cycle (Section 3), soil
carbon stocks (Section 4), and managing soil carbon
(Section 5). The material in these sections is a
composite of the information presented and
discussed at the workshop, including the small
group discussions, and scientific information from
the recent literature. Section 6 outlines and
describes soil carbon management strategies
developed at the workshop for conserving and
sequestering carbon in soils and details specific
management practices that the workshop
participants determined would be important for
achieving the soil management goals. Section 7 is
a synthesis and summary of the workshop and
includes the workshop conclusions. Section 8, the
final section, lists areas (identified at the workshop)
in which further research and data gathering would
lead to improved quantification of global soil carbon
and a better understanding of related processes.
Current estimates of global carbon pools and flaxes in
petagrams of eaiboa indudiag annual increases due to
human activities, (sourx: Houghton el aL 1990)
3. THE GLOBAL CARBON CYCLE
Annually, the burning of fossil fuels releases an
estimated 5 Pg (Pg = petagrams = 1015 grams) of
carbon, primarily as carbon dioxide, to the
atmosphere (Post et al., 1990). The burning of
fossil fuels since the industrial revolution accounts
for most of the increase in atmospheric carbon
dioxide (Houghton et al., 1990). Other human
activities, such as changes in land use and
deforestation, also contribute an estimated 2 Pg of
carbon to the atmosphere annually. At the same
time other "greenhouse" gases (e.g., methane,
nitrous oxide, chlorofluorocarbons) have been
increasing in the atmosphere. Greenhouse gases
transmit incoming solar radiation but partially
absorb the earth's long-wave black-body radiation,
thereby, retaining energy which warms the
atmosphere in proportion to the concentration of
greenhouse gases. If the increasing concentration of
greenhouse gasses goes unchecked, it is likely that
there will be an increase in the earth's mean annual
surface temperature (Houghton et al., 1990). It is
theorized that this warming may alter global
climates causing a change in the timing and
distribution of precipitation. In turn, the
distribution of the earth's cycling carbon could be
changed with the possibility of enormous amounts
of carbon being released into the atmosphere from
terrestrial pools because atmospheric carbon is
linked to carbon in the biosphere through the
carbon cycle. There is also linkage between cycling
carbon and climate because of the relationship of
primary productivity (photosynthesis) to
temperature and precipitation.
2
-------
Global Carbon Pools and Fluxes
At the global-scale, the pools of active, or cycling,
carbon are very large (Figure 1). Recently, Post et
al, (1990) characterized the global cycle in detail
and reported that the atmosphere currently contains
about 750 Pg of carbon, terrestrial systems contain
2,000 Pg, oceans 38,000 Pg, and that the geological
store of recoverable carbon is about 4,000 Pg. The
authors note, however, that there is considerable
uncertainty in these numbers.
GLOBAL RESERVOIRS OF CARBON
(Petagrams of Carbon)
Vegetation 550
Soils 1,500
Atmosphere 750
Oceans 38,000
Recoverable Fossil Fuels 4,000
Source Post et al., 1990
An important feature of the carbon cycle is the
exchange of carbon between pools. Annually, at the
global-scale the natural exchanges of carbon are
very large in comparison to the emission of carbon
from fossil fuel burning. For example, green plants
sequester an estimated 100 - 120 Pg of carbon
through photosynthesis annually. At the same time
through respiration they return about 50% of this
carbon to the atmosphere. The other 50% fixed
through photosynthesis and converted to plant
biomass falls to the ground as litter or enters the
soil by root mortality. Decomposition of plant
residues and soil carbon releases 50 - 60 Pg of
carbon back to the atmosphere. An estimated 100 -
115 Pg of carbon is exchanged between the
atmosphere and oceans annually. As with the
estimates of global carbon pools there are
uncertainties in the estimates of carbon exchange
and it is difficult to balance all the carbon fluxes.
Mitigation
As more carbon is injected into the atmosphere
annually and less and less of it is being removed
(either through photosynthesis or absorbed by the
oceans), concern over the potential for climatic
change grows. Plans to mitigate these gaseous
increases are being developed. In addition to
reducing emissions of carbon dioxide and other
greenhouse gases, some components of the carbon
cycle itself may provide a means to sequester
atmospheric carbon to offset increasing carbon
dioxide, primarily through photosynthesis. It has
been proposed that large-scale reforestation could
sequester a great deal of carbon dioxide over the
next 5 or 6 decades, thus offsetting the magnitude of
the annual increase in atmospheric carbon dioxide.
In addition to sequestering carbon in aboveground
biomass, a significant portion of the carbon fixed
through photosynthesis will be allocated
belowground where it is less susceptible to
oxidation. The modeling work of Prentice and Fung
(1990) indicates that terrestrial plants can sequester
at least 2 Pg of carbon per year, thereby offsetting
fossil fuel emissions. A portion of this will be
sequestered in soils. Soils will have several
functions in biosphere mitigation strategies: they will
be a vegetation rooting medium, a source of
nutrients, and a source of moisture.
Unknowns
The global carbon cycle is not fully understood
(Tans et al., 1990) because of its dynamic nature
and the magnitude of the pools and fluxes between
pools. Even though there have been numerous
studies and reports on the size of global carbon
pools and fluxes, large uncertainties remain in the
generally accepted estimates. One of the key
scientific activities currently underway is the
development of simulation models to project the
effects of shifting carbon stocks and the interaction
with temperature and climate.
4. SOIL CARBON POOLS
Of the 2,000 Pg of carbon in terrestrial systems
(Post et al,, 1990; Houghton and Skole, 1990) an
estimated 1,500 Pg is in soils. This is roughly three
times the amount of carbon in vegetation and two
times the amount of carbon in the atmosphere.
There is also an estimated 800 Pg of inorganic non-
cycling, slowly accumulating carbon in carbonates in
soils. Excluding non-cycling fossil carbon in the
geologic reservoir, soils are the largest terrestrial
reservoir of carbon.
3
-------
Global Distribution of Soil Carbon
Not all soils contain the same amount of
carbon. In fact, the variation in carbon
density (the amount of carbon in the top
meter of soil, or less if the soil is shallower,
corrected for coarse fragments and soil bulk
density ) content can be quite high. Post et
al. (1982) report that soil carbon densities
can range from as low as 1.4 kg carbon m*2
to as high as 72.3 kg carbon m'2 [Table 1],
The lower value is representative of soils in
warm dry deserts and the larger value is
representative of wetland soils that have
accumulated a great deal of carbon. These
two ecosystems represent the extremes in
soil carbon accumulation potential-soil
carbon tends to accumulate under cool wet
conditions but not under warm dry
conditions. Soils have carbon carrying
potential, or maximum amount of carbon
that can accumulate, that is limited by
climate and primary production.
If soils are to be used to sequester
atmospheric carbon it is important to know
what kinds of ecosystems store substantia]
amounts of carbon belowground. Again, the
work of Post et al. (1982) is useful for
characterizing where carbon in soils is
distributed, although published reports by
Schle singer (1984) and Waring and
Schlesinger (1985) could also be used. Post
et al. (1982) report that combined, the
worlds forests (tropical, temperate, and
boreal) cover 30% of earth's land surface
and hold approximately 33%, or 470 Pg, of
the carbon held in soils. In contrast, 16% of
the earth's land surface is deserts which
combined contain 82 Pg of carbon, or about
6% of the global stock of soil carbon.
Tundra and wetland soils cover 9% of the
land and contain 28% of the carbon in soils
(394 Pg). According to Post et al. (1982)
there are 2,1 billion hectares of land under
cultivation, that accounts for 168 Pg of soil
carbon. We know that a portion of the
carbon in these soils will be oxidized and lost to the
atmosphere through common agricultural practices
(Mann, 1986). On the other hand, through
intensive management and implementation of
practices that conserve soil carbon this trend may be
reversed and agricultural soils may be a net sink for
atmospheric carbon. Likewise, through
reforestation or afforestation carbon will be
sequestered in above- and belowground pools.
TABLE 1: DISTRIBUTION OF CARBON IN WORLD SOILS
Source: Post et at,, 1982
:' Life-zone
groups
Tropical Forests
Wet.
Moist
Dry
Veiy Dry
Total
Catlxm Percent of
Density . Area total area
(kgm"3) {jiIQ12^2)
19,1
11.4
9.9
6.1
.'4,1
2.4
3.6
ISA
11.9
Soil
carbon
(xl015g)
783-
60.4
23.8
22.0
1845
Percent of
total soil
: cartoon
131
Temperate Forests
Warm 1.1
•'Cool ' ¦. 12-7
Total
fi.6
3,4
12.0
, 9.3 .
61.1
43.2 •
itoj
IS
Boreal Forests
- Wet'
•. Moist
Total
19.3
11.6
6.9 ¦'
¦ . ; : 4.2 - '
111
8.6
• 133:2
48.7
181.9
13,0
Tropical Woodland
& Savanna 5.4
24.0
18 S
129,6
" - 93
Temperate thorn
steppe
7.6
3.9
3.0
29.6
2.1
Cool temperate
! steppe'.
13.3
9.0
6,9
119.7
8.6
. Tropical desert
bush
2.0
1.2
0.9
."'I' ZA
0,2
Desert
Warm
Cool
•. Boreal •
Total
1.4
9.9
10.2
14.0
4.2
2.0 ¦
• 20.2
IS,6
19.6
41.6
20.4
81.6
5.8
Tundra
21.8 ,
¦ 8.8
6:8
191.8
13.7
Wetlands
72.3 '
2.6'
2.2
' 202.4
• 14.5
Cultivated land
7.9
¦ 212
16.4
1673
. 12.0
Grand Totals
mo
13953
mo
Genesis of Soil Carbon
Post et al, (1990) distinguish two soil carbon pools.
One is litter at the soil surface that is estimated to
contain about 72 Pg of carbon. The other is
referred to as soil organic matter, and contains
about 1,300 Pg of carbon. The annual input to each
of these pools is about 60 Pg of carbon. Because
plant litter is at the soil surface, it is more
4
-------
susceptible to decomposition and accounts for about
42 Pg of the carbon that is returned to the
atmosphere annually through decomposition. Only
about 20 Pg of carbon per year are lost via
decomposition from the larger soil organic matter
pool. Globally, carbon in litter has a retention time
or turnover time of about 1,75 years (size of
pool/annual losses = 72 Fg/42 Pg) and carbon
within the soil profile has a retention time of about
65 years (1,300 Pg/20 Pg). These numbers are
average values because through carbon dating, ages
of soil organic matter approaching 10,000 years have
been reported (Martel and Paul, 1974), Carbon in
wetland soils can have particularly long retention
times of more than 10,000 years (Armentano, 1980)
because inundation prevents rapid aerobic
decomposition of organic deposits.
Fertile soils with climatic conditions appropriate for
plant growth accumulate carbon. Soil organic
matter (carbon) positively influences other soil
characteristics such as consistency, water holding
capacity, and size and distribution of water stable
aggregates. The amount of carbon that accumulates
is related to the inputs of carbon to the soil, either
through litterfall or root turnover, soil moisture and
maximum annual temperatures. Jenkinson et al,
(1987) and Jenkinson (1991) have demonstrated that
increasing the inputs of carbon to soils will increase
the amount of carbon stabilized and retained in soil.
Soils that are wet and cool tend to have higher soil
carbon contents than soils that are moist and warm,
or dry and warm, because litter and soil carbon
decomposition proceed at faster rates at elevated
temperatures. Inundated wetland soils have some
of the highest carbon contents because
decomposition rates are reduced due the anaerobic
conditions created by water saturation. Even
though the primary production of wetland systems
can be quite low, under anoxic conditions
decomposition rates are even lower. Consequently,
carbon accumulates.
For many years there has been a been a debate in
the scientific community regarding the legitimate or
definable pools of soil carbon. It became apparent
at the workshop that this debate has yet to be
settled. In the simplest case there are two kinds of
carbon in soils. Living carbon, which includes roots,
microbes, insects, invertebrates, etc., accounts for
about four percent of total soil carbon (Theng et al.,
1989). The non-living or detrital soil carbon
accounts for about 96 - 98% of soil carbon and
consists of about 20% macroorganic matter and
80% humified (partially decomposed or altered)
material that may less susceptible to oxidation
because of its high carbon to nitrogen ratio.
Soil Carbon Accumulation Rates
One of the important issues in sequestering carbon
in soils is whether or not the rates are sufficient to
sequester enough atmospheric carbon to delay or
prevent global warming. The rate of carbon
accumulation in soils is a function of a variety of
factors, but primary production-photosynthetic
carbon fixation, is of paramount importance.
Without primary production there will be no plant
carbon to sequester belowground. Production is
dependent upon favorable climatic conditions and
fertile soils.
Initially, soils developing in newly formed parent
material lack the biology, physical structure, and
chemistry that are essential for significant primary
production, consequently carbon accumulates slowly.
As the soil becomes more organized, colonized and
structured, productivity accelerates and so does the
soil carbon accumulation rate. The rate eventually
slows for most systems as they near their respective
carbon carrying capacity. Therefore, the soils with
the greatest potential carbon accumulation rates are
likely to be those that are somewhat carbon
depleted, but still retain the requisites (e.g.,
nutrients, biology, structure) for primary production.
The extent of these soils may be great and
therefore, they have the potential to sequester a
substantial amount of atmospheric carbon in the
short-term (50 to 250 years).
Alexander et al. (1989) reported soil carbon
accumulation rates for forested soils in southeastern
Alaska that range from 29 to 113 g carbon m*2 yr"1.
Schlesinger (1991 and this workshop) reported that
the long-term soil carbon accumulation rates for
newly formed land surfaces (e.g., mudflows,
retreating glaciers) is 2.4 g carbon m"2 yr*1. For
soils that have lost some, but not all, of their native
carbon, the accumulation or reaccumulation rate
may be greater. Jenkinson (1991 and Appendix D)
reports that carbon reaccumulation in soils that
were carbon depleted by long-term continuous
agriculture but were abandoned, allowing a mixed
deciduous forest to naturally regenerate, have soil
carbon accumulation rates on the order of 25 - 50
g carbon m"2 yr"1. Ariel Lugo (this workshop)
reported soil carbon accumulation rates in managed
tropical systems as high as 120 g carbon m*2 yr" .
If a soil, having an initial carbon density of 8,000 g
carbon m"2 was brought into agricultural production
with conventional cultivation and loses, on average,
5
-------
40 g carbon m'2yr"1, after 50 years the soil would
have lost 2,000 g carbon m"2 (25% of the initial
carbon). If after 50 years of conventional cultivation
this soil is now managed using conservation tillage
practices that instead of losing soil carbon,
accumulate it at a rate of 10 g carbon m"2^"1, it
would take 200 years to reaccumulate the lost
carbon. This assumes that the accumulation rate
stays constant over the entire 200 years, which is not
likely because accumulation rates tend to slow as
the soil carbon carrying capacity is neared. This
simple example emphasizes the fact that soil carbon
is much easier to lose than it is to accumulate. In
terms of carbon sequestration, it is therefore best to
manage soils to minimize carbon losses rather than
allow the carbon to be lost then to try to restore it.
The current amount of cropland is estimated to be
1.5 billion hectares (World Resources Institute,
1990) and provides an opportunity to implement
management practices on a vast amount of currently
managed land for carbon sequestration in soils.
Additionally, degraded soils could also be managed
to sequester additional carbon. Previous estimates
of the amount of degraded land and the severity of
degradation are highly uncertain. Efforts are
underway, however, to obtain more reliable global
estimates (Oldeman, 1990). When the amount and
condition of this land is know, accurate estimates of
the carbon sequestration potential can be made.
UNEP (1986, UNEP is the United Nations
Environment Program) estimated the amount of
degraded land to be about 2- billion hectares, or
15% of the earth's land surface area. If both the
land currently in cultivation and degraded lands
were managed to accumulate carbon at 2.4 g carbon
m"2 yr'1 they would accumulate only about 0.08 Pg
carbon annually, or about three percent of the
annual atmospheric increment (currently estimated
to be 3 Pg carbon per year). If the average
accumulation rate was 30 g carbon m"2 yr"1, then
the annual accumulation rate would be in excess of
1 Pg of carbon, or a third of the annual atmospheric
increment. The feasibility of sequestering this
amount of carbon in soils needs further evaluation.
Loss of Soil Carbon
Because soil carbon is susceptible to oxidation, soils
axe not permanent repositories for carbon. They
are, however, stable reservoirs for carbon that is in
equilibrium with ambient conditions. Changes in
these conditions can lead to a shift in the amount of
carbon stored, including loss to the atmosphere.
Houghton et at. (1983) reported that land use
changes in the past two centuries have released
more carbon from terrestrial systems, including
soils, than fossil fuel burning during the same
period. These land use changes include forest
harvesting, conversion of forests to agriculture, and
expansion of agriculture to meet growing world food
and fiber demands. One of the concerns associated
with global warming is that even a slight increase in
soil temperature could release a great deal of
carbon to the atmosphere.
Soil carbon levels are maintained by continuous
inputs of new carbon. Tillage practices, such as
plowing, accelerate organic matter oxidation through
the mixing and stirring of the soil. There is rapid
loss of soil carbon in the first 20 to 30 years of
cultivation, with losses of soil carbon ranging from
30 to 50% in the uppermost soil horizons and
somewhat less in the lower horizons (Schlesinger,
1985; Balesdent et al., 1988). Mann (1986) also
reported the rapid initial loss of soil carbon with
cultivation, and also reported that the extent of loss
is related to the starting levels of carbon. Soils
initially high in carbon, lost at least 20% of their
carbon during cultivation, while soils very low in
carbon actually gained some with cultivation. Soil
carbon is also lost by erosional processes. It has
been estimated (this workshop) that up to 50% of
eroded soil carbon is oxidized.
There may be natural limits to the amount of
carbon that can be lost from soils. This limit may
be due to increases in the carbon to nitrogen ratio
(C/N) of the organic residues. Organic matter with
high C/N ratios may be a poor quality substrate for
decomposer organisms (limited nitrogen
availability). Substrate quality may also be related
to the presence of specific organic molecules, such
as phenolics, or to the presence of toxic metals such
as aluminum that decrease the quality of the
material.
Conceptualizing and Quantifying Soil Carbon Pools
To predict how carbon cycles through soils it is
essential to be able to characterize the various
forms of soil carbon. There is a real need to better
understand the dynamic nature (residence times) of
soil carbon and the factors that control the
partitioning between forms. For decades soil
scientists have tried to characterize various carbon
fractions and have generally relied upon a host of
extraction methods to quantify these fractions
(Stevenson and Elliot, 1989). For some applications
the extractable forms of carbon are useful but they
are not useful for describing the dynamic character
of soil carbon.
6
-------
LIVING OHB0N [ DETRITal C*RBON
CO
CO
CO
ABOVEGROUND
UTTEH
CO
MMTtCTtO
SOIL CABBC*
uwseticreo
SOS. CAHBON
BELOW GROUND
tBSSiOMANB LEW5MINS LOSSES
A number of conceptual models have been
developed to depict the dynamic properties of soil
carbon. Parton (Parton et al., 1987 and this
workshop) designates three pools of soil carbon
with increasing resistance to biological
decomposition. One is the "active" pool with a short
turnover time (1 to 5 years). Another is the "slow"
pool with intermediate turnover times (20 - 40
years). Third is the "passive" pool with the longest
turnover times (200 to 1,500 years). Similarly,
Michael Beare (this workshop) has proposed three
soil carbon pools; an "active unprotected" pool, an
"active protected" pool, and a "passive* pool. The
two active pools are associated with soil aggregation
with the unprotected pool of carbon being more
labile than the other two pools,
John Duxbury (this workshop) proposed four pools.
First is an "active" pool, that is readily oxidizable
and is controlled by residue inputs and climate.
Second is a pool of "slowly oxidized carbon" that is
associated with soil macroaggregates. Because
tillage affects the size distribution of aggregates-
tending to decrease the number of macroaggregates,
this pool can be affected by management. Third is
a pool of "very slowly oxidized carbon". This is the
pool of carbon associated with soil microaggregates
and is not likely to be affected by management
unless management somehow completely destroys
soil structure. Fourth, is the "recalcitrant" pool.
This pool of carbon is not accessible to
decomposers and is either intercalated between clay
platelets or adsorbed on platelet surfaces. Because
this fraction relies upon physico-chemical
interactions, it is not easily managed.
There is some uncertainty associated with
operationally defined soil carbon pools. Even
though there is evidence to support each of the
conceptual models of soil carbon proposed above,
direct methods for quantifying these pools are still
being developed. It was generally agreed that there
are at least two distinct pools of soil carbon,
protected (from oxidation) and unprotected and that
further delineations are possible. It was also agreed
that efforts need to be made to arrive at a widely
acceptable and applicable definition of soil carbon
pools that reflects the true dynamic nature of soil
carbon and that these pools must be related to the
processes that are responsible for soil carbon
activity. Methods to measure these pools need to
be developed concomitantly.
Simplified conceptual mode! of caibon pools and Duxes in
terrestrial systems.
Forecasting Soil Carbon Dynamics
It has been. recognized for sometime, and was
reiterated at the workshop, that to successfully
model the flow of carbon in soils, the dynamic
nature of soil carbon has to be reflected in the
designated carbon pools or fractions. Simulation
models of soil carbon dynamics under various
climatic and management regimes can provide very
useful information. The usefulness of the forecasts
will depend upon the amount of realism included in
the model. Equally important is the partitioning
and retention time of the various carbon pools from
which and into which carbon flows. Several
mathematical models have been developed for
predicting the behavior of soil carbon. These
include the models of van Veen et al. (1984),
Jenkinson et al. (1987), and Parton et al. (1988).
Each of these was developed for agricultural
applications and has been used successfully. Bill
Parton (this workshop) used his CENTURY
simulation model to simulate the effects of different
agricultural management practices on soil carbon
demonstrating its utility as a heuristic tool. There
is a need for the development of a model of soil
carbon dynamics in tropical, temperate, and boreal
forests.
Unknowns
Because of the magnitude and diverse nature of
soils there remains some uncertainty in the
estimates of global soil carbon. It is unlikely that
these uncertainties will be reduced or eliminated
without collection of additional data. One of the
weaknesses of existing soil carbon databases is the
lack of information on soil bulk density and the
amount of coarse soil fragments (stones etc.).
Without these, it is not possible to accurately
calculate soil carbon pool sizes (Schlesinger, 1985).
-------
At steady-state the exchange of carbon between
pools is balanced-e.g., terrestrial fixation is equal to
terrestrial oxidation. However, with the injection of
large amounts of carbon into the atmosphere from
fossil fuel burning and deforestation, the exchange
of carbon between pools is unbalanced with yet
unknown consequences. Because of the dynamic
nature of the carbon cycle, the system will readjust
to achieve a new balance. Elevated atmospheric
C02 can cause greater net primary production,
thereby sequestering, through photosynthesis,
additional atmospheric carbon. Because of their
size, the oceans are a likely sink for some of the
fossil fuel C02. A recent analysis by Tans et al.
(1990), however, suggests that the oceans are not as
large of sinks as once believed and that a larger
amount of C02 is absorbed instead by terrestrial
ecosystems including plants and soils. The
magnitude of the carbon that is unaccounted for is
about 2 - 3.4 Pg (Tans et. al., 1990) representing
about 4% of terrestrial net annual primary
production (62 Pg/2.7 Pg) and about 3% of the
annual exchange of carbon between the oceans (100
Pg/2.7 Pg) and the atmosphere. If terrestrial
systems in northern temperate regions are the sink
for this missing carbon, as suggested by Tans et al.
(1990) then it seems that would be possible to
locate the increased carbon sequestration in these
areas. But as Post et al. (1990) point out, current
methods of estimating global carbon pools and
fluxes cannot detect such small annual fluxes. So it
is unlikely that the sink for this carbon will be
explicitly identified. It seems, however, that the
long-term accumulation of 2 - 3.4 Pg per year over
the past 100 years should be simple to locate.
In general, soil carbon accumulates slowly, but
because of past management, soils have lost large
amounts of carbon. If soil temperatures warm
appreciably the net effect is likely to be the efflux of
carbon from the soil. The amount of carbon
released into the atmosphere will depend upon the
extent of warming and the distribution of
precipitation. Modeling is one tool that can be used
to project the effects of various climate change
scenarios on the sequestration of carbon in soils.
There is a need to identify soil carbon pools that
reflect the dynamic nature (retention time) of soil
carbon and to develop direct methods for measuring
these pools. If these can be accomplished, the
predictive capability of currently existing models will
be enhanced.
5. MANAGING SOIL CARBON
Managing the terrestrial biosphere to sequester
additional atmospheric carbon is a potential
measure to aid in mitigating global warming and
climate change. Because soils are the largest
terrestrial reservoir of cycling carbon, a component
of the terrestrial biosphere, and are a potential sink
for atmospheric carbon, a large part of this
workshop was used to consider bow soil
management could lead to increased carbon
sequestration in soils.
TABU! Z SOIL CONDITIONS AND
MANAGEMENT PRACTICES THAT PROMOTE
, CARBON ACCUMULATION IN SOILS,
Soil Change " ¦ Practice , Applications
: Cooler soil MulcJi, shade • All soils
Wetter soil Irrigation Dry regions
Increase Fertility : Fertilizer Most soils
Raise subsoil pH Deep liming Arid subsoils
Reduced aeration limited tillage All soils
¦ (source S.W. Buot and UNEP). •
Workshop participants agreed that soil management
affects the amount of carbon sequestered in soils.
In general, past management was not aimed at
carbon sequestration but often capitalized on the
native fertility of soils. This type of land use, or
management, has led to the loss of massive amounts
of soil carbon in the last century (Houghton et al.,
1983). Houghton and Skole (1990) report that
between 90 and 120 Pg of carbon has been lost to
the atmosphere for the interval between 1850 and
1980 just from changes in land use—a significant
portion of this coming from soils. Data was
presented at the workshop demonstrated that with
proper management the carbon content of many
soils can be increased. An important observation is
that management plays an important role in
determining whether a soil will be a source of
atmospheric carbon or a sink for atmospheric
carbon.
8
-------
Carbon Sequestration in Soils
The objectives of managing soils to sequester
carbon are to increase the amount of carbon
entering the soil and to decrease the amount leaving
either through decomposition or erosion. If soils in
wet cool climates accumulate carbon then it follows
that management practices that produce wetter or
cooler soils will promote the accumulation of
carbon. Stan Buol (see presentation summary in
Appendix B) presented a number of soil conditions
that promote the accumulation of carbon and
management practices that can be used to bring
about the desired changes [Table 2]. Buol's list of
soil conditions that favor carbon accumulation are
shown in the accompanying table.
Carbon Carrying Capacity
Every soil has a carbon carrying capacity that is
determined by soil characteristics, vegetation, and
climate. If left undisturbed in a stable climate, soils
will realize their carbon carrying capacity. If
disturbed, they are likely to loose carbon rapidly.
Worldwide, a large portion of the earth's lands have
been degraded resulting in the loss of soil carbon
(Oldeman et al., 1990). With time and the proper
inputs, the carbon carrying capacity in these soils
may once again be realized. It is also possible to
increase the capacity of some soils to store carbon.
Returning soils to their native carbon carrying
capacity and then increasing that storage capacity
may create considerable sinks for atmospheric
carbon.
Decomposition
Decomposition is a microbially mediated process
that breaks down plant residues and produces C02
as a waste product. Plant residues are the
substrate, or energy source, that fuels
decomposition. Without substrate soil microbial
biomass decreases rapidly. The quality of the
substrate is also important to decomposing
organisms. High C/N ratios, or the presence of
phenolic organic compounds may be indicators of
poor substrate quality. Changes in the composition
of the atmosphere, principally C02, may have direct
effects on the quality of plant residues, and
indirectly affect the decomposition of the residue by
microbes.
In addition to substrate quantity and quality
affecting decomposition, soil temperature and
moisture content also constrain decomposition.
Decomposition is slow in cool wet (low available
oxygen supplies for microbial respiration) soils and
rapid in warm, moist, aerobic soils. In warm and
very dry climates decomposition is slow because soil
microbial activity is limited by droughty conditions.
Temperature Effects
Temperature is an important environmental control
on soil carbon. As temperature goes up so does the
rate of decomposition. That is why cooler soils tend
to have more carbon. It was reported at the
workshop (S.W. Buol) that the crucial element for
carbon sequestration bclowground is the maximum
temperature and not the mean. To manage soils
for maximum carbon storage extremely high
temperatures should be avoided. Mulching or using
vegetation to shade soils are effective methods for
reducing extreme soil temperatures. In terms of
storing soil carbon, dryland fallowing-the practice
of leaving semi-arid agricultural lands bare in
alternate years to accumulate sufficient soil moisture
to grow a crop every other year, should also be
avoided because during the fallow years the
temperatures of the bare soil can be very high. The
use of cover crops or crop rotations may
simultaneously achieve the same soil water
objectives and reduce soil temperatures.
Soil Fertility
Soil fertility' is essential for primary production and
is therefore key to sequestering carbon in soils. To
sequester carbon in soils fertility problems need to
be identified and corrected. Soils that have lost
carbon or that are naturally low in fertility have the
potential to store carbon through improved fertility.
While this is a general guideline there are
circumstances that improving soil fertility could lead
to increased emission of greenhouse gases.
Increased nitrous oxide (N20) emissions from
microbially mediated soil processes (nitrification and
denitirification) may result from the application of
nitrogen fertilizers (Bouwman, 1990; van Breemen
and Feijtel, 1990). Per molecule, the radiative
forcing effect of NjO is about 200 times greater
than C02 (U.S. Congress, 1991).
Another fertilization issue was raised by Dale
Johnson in his presentation. It concerns the long-
term effect of fertilization and carbon sequestration.
Forests tend to be nitrogen limited, i.e., there is a
measurable response to nitrogen fertilization, yet
the response is short-lived. That is, for several
years following nitrogen fertilization forests are
more productive but their productivity soon returns
to that prior to fertilization. This is because
9
-------
nitrogen is lost from the system either through
volatilization or leaching as nitrate. Fertilization
with other nutrients such as phosphorous,
potassium, calcium, or magnesium should have a
much longer lasting effect on system productivity
because these nutrients are relatively immobile. In
forested systems these nutrients are more likely to
have a positive effect on soil carbon because they
cycle in place. Nitrogen in organic forms, either
from municipal wastes or from nitrogen fixing
plants, may have a more permanent effect because
the nitrogen is in a form which is less slowly
released and less likely to be lost from the system.
It is also likely that nitrogen fertilization could
stimulate soil microbes and promote the
decomposition and loss of soil carbon.
Soil Aggregation
The degree of aggregation of soil particles into
larger particles or aggregates, and the stability, of
these aggregates are becoming important indicators
of agricultural soil condition (Beare, this workshop).
Macroaggregates are associated with greater soil
carbon contents. Loss of macroaggregates and an
increase in less stable microaggregates are indicative
of soils that have been degraded through cultivation.
Aggregation is not solely a function of carbon
content. Other factors such as soil particle size
distribution and clay mineralogy also relate to
aggregate formation and stability. Management
practices that result in improved aggregate stability
are likely to be those that also promote the
accumulation of carbon in soils. Michael Beare
(this workshop) also showed that inhibiting soil-
building fungi with fungicides decreased the amount
of water stable aggregates. This emphasizes the
need for the judicious use of chemicals and
fertilizers because of negative effects that could
reduce carbon sequestration in soils.
Agricultural Systems; Temperate and Tropical
The amount of organic carbon in soils is a function
of the quantity and quality of carbon inputs and
subsequent losses through decomposition and
erosion. Soil carbon is also a function of soil
chemical and physical properties. Increasing the
inputs of carbon to soils can increase the amount of
carbon sequestered in soils. Minimising soil mixing
and warming will slow decomposition.
Converstion of native prairie soils and forested soils
to conventional agriculture significantly depletes soil
carbon (Jenny, 1980; Coleman et al., 1984; Mann,
1986). Managing agricultural soils to conserve and
store carbon requires the use of conservation tillage
practices to minimize soil disturbances, and the
incorporation of crop residues into the soil to
increase carbon inputs to the soil. Soil temperature
is positively related to decomposition. The use of
mulches or cover crops reduce soil temperatures,
thereby reducing soil carbon losses.
In general, soils have not been managed to retain or
conserve carbon. Economics, as measured by crop
yields and forest productivity, has been the common
metric driving agricultural and forest system
management, but soils are an important, but often
overlooked, component in the production equation.
In addition to being an important carbon reservoir,
organic carbon in soils is important for maintaining
the productivity of soils. Organic carbon contributes
to soil productivity by promoting soil aggregation;
absorbing and holding water; serving as a natural
reservoir of plant nutrients; minimizing wind and
water erosion; and providing exchange sites for
plant nutrients.
Forested Systems: Temperate and Tropical
Historically, forests have not been managed to store
soil carbon. There are mixed reports in the
literature on the effects of tree harvesting on soil
carbon. In general, when forests are harvested and
the inputs of carbon from vegetation stop, soil
carbon decreases (Edwards and Ross-Todd, 1983;
Houghton et al., 1983). There is usually a lag
between harvesting and a measurable decrease in
soil carbon. Initially, there may be an increase in
microbial biomass associated with the nutrient
inputs from slash but total carbon decreases.
Slash-and-burn agriculture (shifting cultivation) and
the conversion of forests to agriculture dramatically
alter the inputs of carbon to soil. In slash-and-burn
agriculture, cleared lands are farmed for several
years then abandoned. Eventually these lands
naturally return to forests and the net effect on soil
carbon is usually small (Nair, 1984). With
increasing world population there increased
pressure to extend the period of time that these
systems are kept in the cultivation part of the cycle
because the per capita area of land available b
decreasing and can no longer support the number of
people participating in shifting cultivation (U.S.
Congress, 1991). Increasing the frequency and
duration of cultivation, the greater the net loss of
soil carbon. Long-term conversion of forests to
conventional agriculture, as done in the
Southeastern U.S. (Delcourt and Harris, 1980), can
result in the depletion of soil carbon. If cleared
10
-------
lands are turned into pastures the carbon content of
the soils may increase because grasses distribute
carbon deeper into soils because of their rooting
habit. With the loss of the massive amounts of
carbon in aboveground biomass from tree
harvesting the net carbon storage in the system is
decreased.
When removing forest slash by burning, a portion of
the material is converted to charcoal. Charcoal is
found in many soils and is evidence of past fires
(Sanford et al., 1985). Shiffman and Johnson (1989)
report that charcoal is a very stable form of carbon
and may remain unoxidized for thousands of years.
They also estimate that the forest floor in forest
plantations where the slash is burned after
harvesting can be 30% charcoal. Burning slash is
not always the best management practice because
hot burns can damage the soil and slow forest
regeneration. Workshop participants also noted
that fine particles of carbon can be translocated by
the movement of soil water deep into soil profiles.
Although forest management is markedly different
than agricultural management, the principles of
minimizing soil disturbance and taking steps to
lower soil temperatures are still applicable foT
conserving carbon in forested soils. The removal of
forest cover results in increased oxidation of soil
carbon through increased soil temperatures,
particularly in large exposed clear-cuts. Soil
disturbances associated with forest harvesting also
contributes to the net loss of soil carbon.
Implementing management practices that minimize
soil disturbance during forest harvesting and
providing soil mulches following harvesting will
conserve soil carbon in forested soils.
Forest fertilisation is one practice that could
increase the amount of primary production in
forests that could translate into increased
belowground carbon sequestration, Forests are
often growing with some nutrient limitations, such
as nitrogen, and will respond to inputs of limiting
nutrients. For nitrogen the response may be
transient because nitrogen can be lost through
leaching or volatilization. The effects of applying
other nutrients such as phosphorous or potassium
should be long-lasting because these nutrients are;
relatively immobile and remain in the forest system.
A potential source of low-cost forest fertilizer may
be municipal, animal, industrial, or food processing
wastes.
Agroforestry is the deliberate mixture of trees with
crop and animal production systems with the
objective that the net benefit of combining these
two systems will be greater than if the systems were
operated independently (Nair, 1984). The losses of
soil carbon from agroforestry may be less than from
conventional tillage in the tropics. Shading of crops
is one advantage of combining agriculture and
forestry because it lowers temperatures, thus
reducing the potential of" soil organic matter
decomposition.
Unknowns
There are a variety of management options in both
agriculture and forestry that can conserve soil
carbon and may lead to increased belowground
carbon sequestration. It is important to evaluate
each management application in terms of carbon
benefit prior to implementation. For instance,
conservation tillage can lead to increased soil
carbon levels in soils that have been managed with
conventional tillage in the temperate region.
However, if the total carbon cost of conservation
tillage is measured, including the carbon costs of
manufacturing and shipping the additional
herbicides that will be needed for weed control, is
conservation tillage leading to more sequestered
carbon? Questions like this need to answered.
Also what is likely to be the long-term effect of
increased carbon inputs to soils? Will the microbial
biomass sequester essential nutrients when they are
needed by crops? Will yields drop?
Managing soils to store carbon is not necessarily a
"no regrets" environmental policy and will not be
accomplished without some costs. There are
potential negative consequences that should be
considered like the potential release of nitrous
oxide. The release of other gaseous decomposition
products from mulching, such as ethylene, may
affect seed germination. There will be societal costs
as well. Will people be wiling to pay more for food
and wood products because production costs have
increased? If society is unwilling to bear those costs
then managing systems to store carbon is not likely
to succeed.
6. STRATEGIES FOR MANAGING SOIL
CARBON AT THE GLOBAL SCALE
Based upon the size of the soil carbon reservoir and
the ease with which carbon can be lost from soils,
there was a workshop consensus that a proactive
effort is needed to protect and preserve this very
large pool of carbon. Workshop participants also
determined that soils could store additional carbon,
li
-------
but they could not definitively conclude whether or
not the storage of new carbon in soils would be
sufficient to offset atmospheric increases in carbon
or to mitigate global warming. It was debated and
finally concluded that soils should be managed to
optimize carbon sequestration and storage.
SOIL CARBON MANAGEMENT
.STRATEGIES '
• Manage to MAINTAIN the global soil
carbon pool
• Manage to RESTORE carbon to carbon-
depleted soils
• Manage to ENLARGE the global soil
carbon pool
It is important to distinguish between maintaining
soil carbon stocks and managing soils to sequester
additional carbon. Maintaining the integrity and
size of the soil carbon pool is primarily to prevent
soils from being a net source of atmospheric
carbon. Taking steps to preserve soil carbon
implicitly acknowledges that it is more difficult to
restore or raise soil carbon levels..
Workshop participants identified three general soil
carbon management strategies to optimize carbon
sequestration and storage: (1) maintain the global
pool of soil carbon, (2) restore carbon in carbon-
depleted soils, and (3) enlarge the size of the global
soil carbon reservoir. These are described below
and followed by a section describing specific
management practices that the workshop
participants determined would be important for
achieving the soil management goals. Three lists of
management practices were developed, one for each
of the soil carbon management strategies. Tie
practices were ranked by the workshop participants
to delineate the perceived benefit of implementing
these practices. These practices are primarily
intended for use in managed forest and agricultural
systems but may be applicable to other systems. At
the time of the workshop, data were not available to
quantify the amount of carbon that could be
sequestered by implementing these practices.
MAINTAINING THE GLOBAL POOL OF SOIL
CARBON
Because soils are the largest terrestrial pool of
active carbon and because a large portion of this
carbon is potentially released to the atmosphere
through human activities, workshop participants
recommended that management practices be
implemented to preserve the size and integrity of
the pool. Managing soils to maintain their current
carbon levels recognizes that soil carbon is labile
and is lost to the atmosphere relatively easily when
the soil is mixed and stirred as in conventional
agricultural systems. It also recognizes that it takes
more time to raise soil carbon levels than it does to
lose soil carbon. Workshop participants concluded
that it is better to protect soil carbon before it is
lost, than it is to allow it to be lost and then try to
restore it. A significant loss of soil carbon to the
atmosphere could exacerbate global warming.
Management Practices to Maintain Soil Carbon
Levels
The management practices described below were
identified by the workshop participants as those that
could be used to maintain the amount of carbon in
soils.
• Maintain Soil Fertility: Sustained inputs of
carbon to soils from primary production are
essential for maintaining current levels of soil
carbon. Interruption of these carbon inputs
results in loss of soil carbon by the
readjustment of the soil system to a lower
equilibrium level of soil carbon. Maintaining
soil fertility helps sustain primary production,
which in turn helps to sustain the input of
carbon belowground and the sequestration of
carbon in soils. [High Priority]
• Concentrate Tropical Agriculture:
Concentrating or intensifying tropical
agriculture is aimed at preventing tropical
deforestation and the soil carbon loss associated
with it, and thereby maintaining the existing
stocks of soil carbon. Concentrating tropical
agriculture is the practice of directing resources
and management for use only on the best
agricultural lands. The intended outcome is
that less land will be needed to produce the
same, if not more, volume of agricultural
products. In achieving this, the need for slash-
and-burn agriculture (deforestation) is
diminished and the carbon losses associated
with it. [High Priority]
12
-------
• Preserve Natural Wetlands: The.
inundated soils in wetlands contain
a great deal of carbon that has a
very long retention time if
undraincd. The practice of draining
these systems, primarily for agrarian
purposes, results in the rapid
oxidation and loss of carbon to the
atmosphere. Preserving wetlands
insures the carbon storage function
of these lands. [High Priority]
• Increase Efficiency of Forest
Product Use.* Increasing the
efficiency of forest harvesting (by
reducing waste) and increasing the
life-span of forest products
(including recycling) is aimed at
reducing the demand for forest
harvesting. Reducing forest
harvesting, or increasing the interval
between harvests, will help to
maintain the large above- and
belowground carbon stocks in the
world's forests. [High Priority]
Agricultural and Forestry Practices to
Maintain Soil Carbon
Management Practice
Maintain Soil Fertility
Concentrate Tropical Agriculture
Preserve Natural Wetlands
Increase Efficiency of Forest Product Use
Retain Forest Slash on Site
Minimize Site Disturbance
Use Prescribed Burns to Maximize C Storage
Control Erosion
Mulching
Leave Crop Residues
Minimum Tillage
Incorporate Crop Residues ;
Priority*
H
:H ^ '
H
M '
M '
. : m
M
M
M
M
•: • l
• Conservation Tillage: Conservation
tillage is the agricultural practice of
retaining crop residues on-site and
reducing the number or severity of
soil physical manipulations used to
prepare seedbeds, control weeds, and apply
fertilizer, thereby diminishing the loss of soil
carbon as compared to conventional tillage
practices that result in the loss of soil carbon.
Conservation tillage includes the practices of
either minimum tillage or no-tillage and
farming along the contour of the land.
[Medium Priority]
• Retain Forest Slash on Site: Forest harvest
residues are an important source of nutrients
for successive rotations as well as an important
post-harvest source of soil carbon. Often these
residues are either removed or burned to
expedite and reduce the cost of replanting.
Removing slash removes nutrients that help to
get the next rotation of trees established.
Burning may cause nitrogen to be lost from the
system by volatilization. If hot enough, burning
can damage the soil resulting in future
decreases in system productivity. Burning may
also leave the soil bare, making it more
susceptible to erosion, particularly in humid
regions on steep slopes. Leaving residues on
site may make it more difficult and more
^Relative Priority: H * High, M « Medium, L = Low
expensive to replant, but because of increased
nutrient availability, lower soil temperatures,
and increased water availability, rotations may
become established more quickly and be more
productive while conserving soil carbon.
[Medium Priority]
Minimize Site Disturbance: Extensive use of
ground systems during forest harvesting may
compact or disrupt large portions of the soil in
the harvest areas. Likewise, yarding or skidding
logs to landings can compact and expose the
mineral soil. This can reduce the productivity
of the site and promote the loss of soil carbon
by oxidation or through erosion. Harvest and
management practices that minimize or
eliminate site disturbance will help maintain the
productivity of the forest system and conserve
soil carbon. [Medium Priority]
Use Prescribed Burning to Maintain Carbon
Storage in Soils: Burning has been used for
centuries as an effective method for removing
slash following a forest harvest or in shifting
agriculture. Hot slash burns (i,e., hot dry,
13
-------
¦windy weather with dry slash) may severely
damage the soil and affect the productivity of
the system. These burns can even combust soil
carbon and should be avoided. If burning is
required, lighter burns (i.e., in cool wet
weather) will be less damaging overall to the
system and make it easier for the ecosystem to
transition through harvesting to a new forest.
Changing the way slash is burned may, in some
situations, increase rather than decrease, the
potential for a site to sequester and store
carbon. Burning will convert some of the slash
to charcoal, effectively sequestering the carbon
for thousands of years. [Medium Priority]
• Control Erosion: Controlling soil erosion is
intended to protect the soil resource. The
surface layer of soil or topsoil is generally more
fertile and has better water holding
characteristics than the lower soil horizons.
Because of its proximity to wind and rain,
topsoil is the most likely part of the soil to be
physically eroded. Physical loss or degradation
of the soil resource diminishes primary
production and consequently, carbon
sequestration belowground. [Medium Priority]
• Mulching: The rate of, soil organic matter
decomposition is positively related to soil
temperature. Mulching or using plant residues
to cover the soil reduces extreme soil
temperatures, thereby slowing decomposition,
resulting in the retention of more carbon in the
soil. [Medium Priority]
• Leave Crop Residues: Crop residues are an
important source of carbon and nutrients in
agricultural systems. Leaving crop residues
helps to maintain soil carbon levels by
maintaining the input of carbon to the soil and
by serving as a mulch that reduces soil
temperatures, [Medium Priority]
• Incorporate Crop Residues: Incorporating
crop residues into the soil slows its
decomposition and insures that more of the
residue decomposition products remain in the
soil, thereby helping to maintain soil carbon
levels in agricultural soils. Residue
incorporation improves soil water holding and
infiltration characteristics, and in the tropics,
prevents insects from removing the residue.
[Low Priority]
RESTORING SOIL CARBON IN CARBON-
DEPLETED SOILS
An opportunity exists to restore carbon in soils that
are depleted in carbon due to management.
Worldwide, there may be millions of hectares of
unproductive forest land or abandoned agricultural
lands that could store additional carbon.
Management practices that bring these once
productive lands back into primary production will
lead to increased carbon sequestration above- and
belowground. Soils that are the most carbon
depleted will require more inputs to bring back
their carbon sequestration potential. The long-term
objective of this strategy is to restore soil carbon to
previous levels.
Management Practices to Restore Carbon in
Carbon-Depleted Soils
It was the consensus of the workshop participants
that soils somewhat depleted in carbon offer the
greatest potential for sequestering additional carbon
because their carbon accumulation rates are much
faster than soils approaching their carbon carrying
capacity. Worldwide, the extent of soils that have
lost carbon appears to be great (Oldeman et al.,
1990). The management practices described below
were identified as those that would be useful for
restoring soil carbon to previous levels. The
amount of carbon that can be sequestered by
implementing these is, however, currently unknown.
• Reforestation: Large-scale reforestation has
the potential to sequester and store large
amounts of new carbon both above- and
belowground, particularly on carbon-depleted
soils. Some deforested areas will regenerate
naturally while others will only require new tree
planting. Others will require intensive
management and large inputs of resources.
[High Priority]
• Improve Soil Fertility: Soils that are carbon
depleted are also likely to lack some other soil
nutrients, which may limit primary production.
Improving the fertility of these soils to support
vegetation is key to restoring carbon in these
soils. [High Priority]
• Use Municipal, Animal, Industrial and Food
Processing Wastes: Restoring carbon in
carbon-depleted soils may be limited by the
lack of plant nutrients needed for primary
14
I
-------
production or by the cost of
fertilizer to replenish nutrients.
Municipal, animal, industrial,
and/or food processing wastes may
be ideal sources of low-cost forest
fertilizers. Using them in forests or
in agriculture, particularly on
marginal lands, can provide water
and essential nutrients to vegetation,
thereby promoting a higher level of
primary production and above- and
belowground carbon sequestration.
[High Priority]
Agricultural and Forestry Practices for
Restoring Carbon in Carbon-
Depleted Soil
• Concentrate Tropical Agriculture:
Concentrating or intensifying
tropical agriculture is aimed at
preventing tropical deforestation
and the soil carbon loss associated
with it, and thereby maintaining the
existing stocks of soil carbon.
Concentrating tropical agriculture is
the practice of directing resources
and management for use only on
the best agricultural lands. The
intended outcome is that less land
will be needed to produce the same,
if not more, volume of agricultural
products. In achieving this, the
need for slash-and-burn agriculture
(deforestation) is diminished and the carbon
losses associated with it. The low-productivity
lands removed from shifting agriculture by
concentrating tropical agriculture can be
reforested or revegetated, thus sequestering
carbon in biomass and initiating the natural
process of restoring soil carbon. [Medium
Priority]
• Remove Marginal Lands from Intensive
Agricultural Production: Marginal lands are
usually steep and prone to erosion. They also
tend to be naturally infertile and are not well
suited for agricultural production. With low-
cost fuel and fertilizer they have been brought
into agricultural production, but cultivation of
these lands has resulted in depletion of soil
carbon stocks. By removing these lands from
intensive agricultural production, they can be
reforested or revegetated to sequester carbon in
above- and belowground biomass and restore
soil carbon to previous or near previous levels.
[Medium Priority]
• Control Erosion: Physical loss or degradation
of the soil resource diminishes primary
Management Practice
Reforestation
JMaintain or Improve Soil Fertility
Use Municipal, Animal, Industrial and
Food Processing Wastes As a Source
of Low-Cost Fertilizer
Concentrate Tropical Agriculture
Remove Marginal Lands From Intensive
Agricultural Production
Control Erosion
Urban Forestry
Priority*
H
':H;
v. M
M
M
¦ L;
] aRelative Priority: H = High, M = Medium, L = Low
production and consequently, carbon
sequestration belowground. By controlling soil
erosion and preventing loss of the soil resource,
they can be used to support vegetation and
sequester carbon belowground. [Medium
Priority]
Urban Forestry: An opportunity exists, albeit
small, in urban areas to use small forests and
individual trees to capture and store carbon in
above and belowground biomass and as soil
carbon. The objective is to restore soil carbon
that has been lost because of the urban land
use. Additional societal benefits of urban
forestry include: recreation, lower urban
temperatures, and air purification. Lower
urban temperatures during the summer months
will reduce air-conditioning usage, thereby
conserving the carton that would otherwise be
used to run urban air-conditioners, [Low
Priority]
15
-------
ENLARGING THE SIZE OF THE
GLOBAL SOIL CARBON RESERVOIR
The objective of this strategy is to
manage forested and agricultural
systems in ways that increase their
productivity and their allocation and
storage of carbon belowground. The
essential aim of this strategy is to
increase the carbon carrying capacity of
soils. Most of the opportunities to
enlarge this pool may be in agriculture
because agricultural systems are
generally more intensely managed than
forested systems. Workshop participants
surmised that the marginal return,
measured in terms of stored carbon,
would probably be greater by
implementing management practices
aimed at "maintaining" and "restoring*
soil carbon than by attempting to
enlarge the global pool of soil carbon.
Management Practices for Enlarging
the Global Soil Carbon Reservoir
Agricultural and Forestry Practices for
Enlarging the Global Pool
of Soil Carbon
Management Practice
Minimum Tillage
Improve So'd Fertility
Concentrate Tropical Agriculture \
Minimize Dryland Fallowing .
Retain Forest Slash on Site
Leave Crop Residues
Incorporate Crop Residues
Remove Marginal Lands From Intensive
Agricultural Production
Use Municipal, Animal, Industrial and
Food Processing Wastes as a Source
of Low-Cost Fertilizer
Conservation Tillage; In the
context of enlarging global soil
carbon stocks, widespread
implementation of conservation
tillage practices could lead to
additional sequestration of carbon
in soils. Implementing conservation
tillage practices implies that the use
conventional tillage practices will decline, thus
reducing the oxidative loss of soil carbon.
[High Priority]
Improve Soil Fertility: Lack of nutrients often
limits primary production. Eliminating or
reducing nutrient limitations will improve
primary production leading to greater
sequestration of carbon in soil. [High Priority]
Concentrate Tropical Agriculture: By
concentrating tropical agriculture, lands
removed from shifting agriculture can be
reforested or revegetated, leading to increased
soil carbon stocks. [High Priority]
Minimize Dryland Fallowing: Fallowing is the
practice of leaving semi-arid agricultural lands
bare in alternate years to accumulate sufficient
soil moisture to grow a crop every other year.
Dryland fallowing is a widely used agricultural
.Mulching
Control Erosion
Priority*
; H
H
;
H ^
M
M
M
¦ • M
M
M
aRe1aiivc Priority. H = High, M - Medium, L - Low
practice. There is gathering evidence that this
practice promotes the loss of soil organic
carbon because the soil is usually kept bare by
mechanical means. This stirring and mixing of
the soil combined with increased soil
temperatures, due to lack of cover, results in
rapid and thorough oxidation of organic carbon.
Use of cover crops or crop rotations may
achieve the same soil water objectives and
eliminate the need to fallow, [High Priority]
Retain Forest Slash on Site: In the context of
enlarging stocks of soil carbon, retaining forest
slash on site following forest harvesting is
aimed at retaining nutrients and water to
promote the re-establishment of new forest
vegetation as quickly as possible. If the cycle of
burning or removing residues is interrupted, soil
carbon stocks will not only be maintained but
potentially increased. [Medium Priority]
16
-------
• Leave Crop Residues: Crop residues are an
important source of carbon and nutrients in
agricultural systems. Leaving crop residues may
help to increase soil carbon levels by increasing
the input of carbon to the soil and by serving as
a mulch that reduces soil temperatures. This is
an important component of conservation tillage.
[Medium Priority]
• Incorporate Crop Residues: Incorporating
crop residues into the soil slows its
decomposition and insures that more of the
residue decomposition products remain in the
soil thereby help to maintain soil carbon levels
in agricultural soils. Residue incorporation
improves soil water holding and infiltration
characteristics, and in the tropics, prevents
insects from removing the residue. [Medium
Priority]
• Remove Marginal Lands from Intensive
Agricultural Production: Marginal lands are
usually steep and prone to erosion. They also
tend to be naturally infertile and are not well
suited for agricultural production. With low-
cost fuel and fertilizer they have been brought
into agricultural production, but cultivation of
these lands has resulted in depletion of soil
carbon stocks. By removing these lands from
intensive agricultural production, they can be
reforested or revegetated to sequester carbon in
above- and belowground biomass, restore soil
carbon to previous or near previous levels and
potentially increase global soil carbon stocks.
[Medium Priority]
• Use Municipal, Animal, Industrial and Food
Processing Wastes: Restoring carbon in
carbon-depleted soils may be limited by the
lack of plant nutrients needed for primary
production or by the cost of fertilizer to
replenish nutrients. Municipal, animal,
industrial, and/or food processing wastes may
be ideal sources of low-cost forest fertilizers.
Using them in forests or in agriculture,
particularly on marginal lands, can provide
water and essential nutrients to vegetation,
thereby promoting a higher level of primary
production and above- and belowground carbon
sequestration. [Medium Priority]
• Mulching: The rate of soil organic matter
decomposition is positively related to soil
temperature. Mulching or using plant residues
to covering the soil reduces extreme soil
temperatures, thereby slowing decomposition,
resulting in the retention of more carbon in the
soil. [Medium Priority]
• Control Erosion; Controlling soil erosion is
intended to protect the soil resource. The
surface layer of soil or topsoil is generally more
fertile and has better water holding
characteristics than the lower soil horizons.
Because of its proximity to wind and rain,
topsoil is the most likely part of the soil to be
physically eroded. Physical loss or degradation
of the soil resource diminishes primary
production and consequently, carbon
sequestration belowground. [Low Priority]
7. WORKSHOP SUMMARY AND
CONCLUSIONS
This workshop was an excellent forum for a
scientific debate on the potential of soils to
sequester additional carbon from the atmosphere.
Two primary conclusions can be drawn from the
workshop. First, that steps should be taken to
protect and preserve the size and intejp*ity of the
global reservoir of soil carbon because continued
losses of soil carbon to the atmosphere could
exacerbate global warming and climatic change.
Second, that steps should be taken to manage soils
and ecosystems to store additional carbon. The
latter being accomplished predominately by
increasing net primary production. The major
uncertainties related to carbon sequestration in soils
and specific strategies for addressing these
uncertainties and for managing soils to store carbon
were also identified at this workshop.
Major conclusions of the workshop participants:
• Soils are an important component of the global
carbon cycle, containing a large pool of active,
cycling carbon. As such, they are a very large
potential source of atmospheric carbon, but
they also represent a large potential sink for
carbon, if managed properly.
• Uncertainties exist in the success of widespread
implementation of soil management practices
because of the potential for the occurrence of
concomitant negative effects. At some locations
the associated negative effects may outweigh
the positive benefits. For instance, under
certain circumstances some of these practices
could lead to the emission of gases (e.g., nitrous
oxide) that have a greater radiative forcing than
carbon dioxide. In terms of global warming,
17
-------
this scenario would be counterproductive. The
implementation of any new or altered
management practices should be considered on
a site-by-site or region-by-region basis prior to
implementation, and should be evaluated in
terms of the effect on carbon pools and fluxes.
• Three strategies for managing soil carbon are
proposed: (1) manage soils to maintain current
levels of soil carbon, (2) manage carbon
depleted soils to restore carbon to former
levels, and (3) manage soils to enhance the size
of current soil carbon pools. For each of these
a variety of opportunities exist for capturing
atmospheric carbon, via photosynthesis, and
storing it in soils and aboveground biomass.
• In addition to storing carbon assimilated from
the atmosphere, managing soils to conserve
carbon will have other benefits. These include:
(1) increased soil water holding capacity, (2)
increased nutrient availability, (3) improved soil
physical properties, and (4) decreased soil
erosion by wind and water. Together these
should lead to (5) increased food and fiber
production. Managing soils to conserve carbon
will help to produce more sustainable forest
and agricultural systems.
The consensus on the initial workshop question
"Can soils be used to store sufficient carbon to aid
in mitigating global climate change?" is that we need
more reliable information to provide a definite
answer.
8. RESEARCH RECOMMENDATIONS
Throughout this workshop, the lack of sufficient,
reliable, quantitative data, and numerous areas of
uncertainty related to soil carbon were identified.
Research and data gathering in identified areas will
improve our quantification of global soil carbon and
related processes. Here we list those topics thought
to be critical for a more complete analysis of the
relationship between soils and global climatic
change. This list, although not exhaustive, provides
guidance for research in areas that could provide
valuable information or tools for evaluating the role
of soils in the global carbon cycle with a focus on
the potential of soils to sequester and store
additional carbon from the atmosphere to mitigate
the effects of global warming.
• Define soil carbon pools based upon lability and
soil processes, and determine the factors that
control the partitioning of carbon into the
respective pools.
• Identify and characterize the specific fractions
of soil carbon that are manageable and
practices that are effective for managing them.
• Improve estimates of global soil carbon by
conducting large-scale statistically designed soil
surveys coupled with intensive soil sampling and
physical and chemical analysis.
• Quantify above- and belowground carbon pools
and fluxes in specific ecosystems for the
purpose of developing general principles of
ecosystem carbon dynamics from specific
examples.
• Characterize and quantify the factors that
control soil carbon fluxes.
• Develop soil carbon methods for characterizing
and quantifying the true size and lability of soil
carbon pools.
• Characterize and quantify the role of abiotic
soil factors in stabilizing soil carbon.
• Quantify the effects of land use and
management, including agricultural and
forestry, on soil carbon.
• Conduct experiments to characterize the effects
of altered climatic conditions on terrestrial
plant carbon fixation and allocation, focusing on
the quantity and quality of carbon in detritus
and in belowground allocation.
• Develop simulation models that accurately
project how carbon fluxes (and thus pools and
feedbacks to the atmosphere) will shift in
specific ecosystems under a series of altered
climate scenarios.
• Quantify the economics of implementing soil
management practices that sequester and store
carbon.
LITERATURE CITED
Alexander, E.B., E. Kissinger, R.H. Huecker, and P.
Cullen. 1989. Soils of southeast Alaska as sinks for
organic carbon fixed from atmospheric carbon-
dioxide. p.203-210. Jn E.B. Alexander (ed.)
Proceedings of Watershed '89; A conference on the
18
-------
stewardship of soil, air, and water resources,
USDA Forest Service, Alaska Region, R10-MB-77
Armentano, T.V. (ed). 1980. The role of organic
soils in the world carbon cycle-problem analysis
and research needs. Report to the U.S.
Department of Energy No.CONF-7905135. The
Institute of Ecology, Indianapolis, Indiana.
Balesdent, J., G.H. Wager, and A. Mariotti. 1988.
Soil organic matter turnover in long-term field
experiments as revealed by carbon-13 natural
abundance. Soil Sci. Soc. Amer. I, 52:118-124.
Bouwman, A.F. 1990. Exchange of greenhouse
gases between terrestrial ecosystems and the
atmosphere, p.62-127. In A.F. Bouwman (ed.) Soils
and the greenhouse effect; Proceedings of the
international conference on soils and the
greenhouse effect, John Wiley and Sons, New York.
Coleman, D.C., C.V. Cole, and E.T. Elliott. 1984,
Decomposition, organic matter turnover and
nutrient dynamics in agroecosystems, p. 83-104. Jn
R. Lowrance, B.R. Stinner, and G J. House (eds,).
John Wiley and Sons, New York.
Delcourt, H.R., and W.F. Harris. 1980. Carbon
budget of the southeastern U.S. biota: Analysis of
historical change in trend from source to sink.
Science 210:321-323.
Edwards, N.T., and B.M. Ross-Todd. 1983. Soil
carbon dynamics in a mixed deciuous forest
following clear-cutting with and without residue
removal. Soil Sci. Soc. Am. J, 47:1014-1021.
Houghton, J.T., GJ. Jenkins, and JJ. Ephraums
(eds.). 1990. Policymakers Summary, pjt-xxxiii. la
Climate Change: The IPCC Scientific Assessment.
Cambridge University Press, Cambridge, England.
Houghton, RA, J.E. Hobbie, J.M. Melillo, B.
More, BJ. Peterson, G.R, Shaver, and G.M.
Woodwell, 1983. Changes in the carbon content of
terrestrial biota and soils between 1860 and 1980: A
net release of C02 to the atmosphere. Ecological
Monographs 53:235-262.
Houghton, RA., and D.L. Skole. 1990. The long-
term flux of carbon between terrestrial ecosystems
and the atmosphere as a result of changes in land
use. Research Project of the Month - July. Carbon
Dioxide Research Program, Office of Health and
Environmental Research, U.S. Department of
Energy.
Jenkinson, D.S., P.B.S. Hart, J.H, Rayner, and L.C,
Parry. 1987. Modelling the turnover of organic
matter in long-term experiments at Rothamsted.
INTECOL Bulletin 15:1-8.
Jenkinson, D.S. 1991. The Rothamsted long-term
experiments: are they still of use? Agron, J, 83:2-
10.
Jenny, H. 1980. The Soil Resource: Origin and
Behavior. Springer-Verlag, New York.
Mann, L.K 1986. Changes in soil carbon storage
after cultivation. Soil Sci. 142:279-288.
Martel, YA„ and EA. Paul. 1974. The use of
radiocarbon dating of organic matter in the study of
soil genesis. Soil Sci. Soc, Amer. J. 38:501-506.
Nair, PJC.R. 1984. Soil productivity aspects of
agroforestry. International Council for Research in
Agroforestry, Nairobi, Kenya.
Oldeman, L.R., R.TA. Hakkeling, and W.G,
Sombroek. 1990. World map of the status of
human-induced soil degradation: An explanatory
note. International Soil Reference and Information
Center, Wageningen, The Netherlands.
Parton, WJ., D.S, Schimel, C.V, Cole, and D.S,
Ojima, 1987. Analysis of factors controlling soil
organic matter levels in great plains grasslands. Soil
Sci. Soc. Am. J. 51:1173-1179.
Parton, WJ., J,W.B. Stewart, and C.V. Cole. 1988.
Dynamics of C, N, P and S in grassland soils: a
model. Biogcochem. 5:109-131.
Post, W.M., W.R. Emanuel, PJ. Zinke, and A.G.
Stangenberger. 1982. Soil carbon pools and world
life zones. Nature 298:156-159,
Post, W.M., T.-H. Peng, W.R, Emanuel, A.W,
King, V.H. Dale, and D.L. DeAngelis. 1990. The
global carbon cycle. American Scientist 78:310-326.
Schlesinger, W.H. 1984. Soil organic matter: a
source of atmospheric C02- 1984. p, 111-127. In
G.M. Woodwell (ed.) The role of terrestrial
vegetation in the global carbon cycle; Measurement
by remote sensing. SCOPE 1984, John Wiley &
Sons Ltd,
Schlesinger., W.H. 1985. Changes in soil carbon
storage and associated properties with disturbance
19
-------
and recovery, p,194-220, jn J.R. Trabalka and D.E,
Reichle (eds.) The changing carbon cycle: A global
analysis. Springer-Verlag, New York.
Waring, R.H., and W.H. Schlesinger. 1985. Forest
ecosystems: Concepts and management. Academic
Press Inc., San Diego, California.
Schlesinger, W.H. 1990. Evidence from
chronosequence studies for a low carbon-storage
potential.of soils. Nature 348:232-234.
Stevenson, FJ., and E.T. Elliot. 1989.
Methodologies for assessing the quantity and quality
of soil organic matter. In D.C. Coleman, J.M,
Oades, and G. Uehara (eds.) Dynamics of soil
organic matter in tropical ecosystems. University of
Hawaii Press, Honolulu, Hawaii.
Sanford, R.L., I. Saldarriaga, K.E. Clark, C. Uhl,
and R. Herrera. 1985. Amazon rain-forest fires.
Science 227:53-55.
Schiffman, P.M., and W.C. Johnson. 1989.
Phytomass and detrital carbon storage during forest
regrowth in the southeastern United States
Piedmont. Can. J. For. Res. 19:69-78.
Tans, P.P., I.Y. Fung, and T. Takahashi. 1990.
Observational constraints on the global atmospheric
C02 budget. Science 247:1431 - 1438.
Theng, B.K.G., K.R. Tate, and P. Sollins. 1989.
Constituents of organic matter in temperate and
tropical soils. In D.C. Coleman, J.M. Oades, and
G. Uehara (eds.) Dynamics of soil organic matter
in tropical ecosystems. University of Hawaii Press,
Honolulu, Hawaii.
UNEP. 1986. Farming systems principals for
improved food production and the control of soil
degradation in arid and semi-arid tropics.
ICR IS AT, Hyderabad, India.
U.S. Congress. 1991. Changing by degrees: Steps
to reduce greenhouse gases. Office of Technology
Assessment, OTA-O-482 Washington, D.C.
Van Breemen, N. and T.CJ. Feijtel. 1990. Soil
processes and properties involved in the production
of greenhouse gases, with special relevance to soil
taxonomic systems, p.195-223. In A.F. Bouwman
(ed.) Soils and the greenhouse effect: Proceedings
of the international conference on soils and the
greeenhouse effect, John Wiley and Sons, New
York.
Van Veen, J A., J.N. Ladd, MJ. Frissel. 1984.
Modelling C and N turnover through the microbial
biomass in soil. Plant and Soil 76:257-274.
20
-------
APPENDIX A; WORKSHOP AGENDA
SEQUESTERING CARBON IN SOILS; A WORKSHOP TO
EXPLORE THE POTENTIAL FOR MITIGATING GLOBAL CLIMATE CHANGE
Final Agenda
Time Monday, February 26.1990
0830 Agriculture and Forestry Field Tour - Meet at Neodel's Iim Parking Lot
1700 Return to EPA Laboratory
Tuesday. February 27. 1990
Room 211 - Oregon State University Memorial Union
0830 Welcome
Tom Murphy, Director
EPA Environmental Research Laboratory
Corvallis, OR
0840 Workshop Objectives and Procedures
Mark Johnson
EPA Environmental Research Laboratory
Corvallis, OR
0850 Introductions
Plenary Session #1 - Carbon Cycles
0915 Soil Genesis and the Carbon Cycle
Stan Buol
Department of Soil Science
North Carolina State University
Raleigh, NC
0940 Perspectives on the Global Carbon Cycle
William Schlesinger
Department of Botany
Duke University
Durham, NC
1005 Perspectives on the Terrestrial Carbon Cycle
Richard Houghton
The Ecosystems Center
Marine Biological Laboratory
Woods Hole, MA
21
-------
Time
1030 BREAK
1045 Small Working Groups - Session #1
1145 Small Working Group Reports
1200 Lunch
Plenary Session #2 - Soil Carbon
1300 Soil Carbon in Forested Ecosystems #1
Ariel Lugo
Institute of Tropical Forestry
Rio Piedras, PR
1325 Soil Carbon in Forested Ecosystems #2
Phil Sollins
Department of Forest Science
Forest Sciences Laboratory
Oregon State University
Corvallis, OR
1350 Soil Carbon in Agroecosysterns #1
John Duxbury
Department of Agronomy
Cornell University
Ithaca, NY
1415 Soil Carbon in Agroeeosystems #2
Michael Beare
Institute of Ecology
University of Georgia
Athens, GA
1440 Soil Carbon in other Ecosystems
Mac Post
Environmental Sciences Division
Oak Ridge National Laboratory
Oak Ridge, TN
1505 BREAK
1520 Small Working Groups - Session #2
1645 Small Working Group Reports
1700 ADJOURN
22
-------
Time
Workshop Banquet
Peavey Lodge, OSU Peavey Arboretum
1900 DINNER
2030 The Role of Scientists in Developing Environmental Policy*
David Bella, Professor
Department of Civil Engineering
Oregon State University
Corvallis, OR
2300 ADJOURN
Wednesday. February 28. 1990
Room 211 - Oregon State University Memorial Union
Plenary Session #3 - Managing Soil Carbon
0830 Introductory Remarks
Mark Johnson
0845 Managing Soil Carbon in Tropical Agroecosystems
Rattan Lai
Department of Agronomy
The Ohio State University
Columbus, OH
0910 Managing Soil Carbon in .Agroecosystems #1
William Parton
Natural Resource Ecology Laboratory
Colorado State University
Fort Collins, CO
0935 Managing Soil Carbon in Agroecosystems #2
Paul Rassmussen
USDA-ARS and Oregon State University
Columbia Basin Agricultural Research Center
Pendleton, OR
1000 BREAK
23
-------
Time
1015 Managing Carbon in Forested Ecosystems
Kermit Cromaek
Department of Forest Science
Forest Sciences Laboratory
Oregon State University
Corvallis, OR
1040 Managing Soil Carbon in Forested Ecosystems
Dale Johnson
Desert Research Institute
University of Nevada
Biological Sciences Center
Reno, NV
1105 Small Working Groups - Session #3
1200 LUNCH
1300 Small Working Groups - Session #4
1430 Small Working Group Reports
1500 Group Discussion
1600 ADJOURN
24
-------
APPENDIX B: PRESENTATION SUMMARIES
This section contains short summaries of each of the scientific presentations. These summaries highlight the
salient points that each speaker made. They were compiled by designated rapporteurs or by the presenter.
Several presenters submitted written papers to further document their presentations. The submitted papers
follow in Appendix C.
Stan Buol, Department of Soil Science, North Carolina State University, Raleigh, NC
1. From a soil genesis perspective, the two forms of soil carbon of interest are the carbonate form (which is
not discussed here) and organic carbon (OC) in the soil. The difference between soil and geologic material
is the contribution of living things. Organic matter is greatly influenced by the type of mineral material that
it is growing in.
2. The most difficult and unknown questions are just what is soil carbon, what is its composition, and how does
it vary in behavior throughout the world? Most of the soil carbon research to date has looked at how to
extract carbon." This research does not address how it behaves in the field because some forms of carbon
may provide no energy for microbial respiration.
3. A simplified view of soil carbon is that some of the carbon is leached, some is oxidized to C02, and some
goes through hum ideation reactions. Carbon dissolved in groundwater is readily observable in sandy, humid
areas where there are black rivers (i.e., Rio Negro, Brazil). This leaching is part of the podzolization process
which forms spodic horizons, but some of the carbon is leached completed out of the soil. Organic matter
near the surface oxidizes rapidly to C02. Organic matter in the soil undergoes a series of reactions in the
humification chain. In the process some carbon is released as C02 and some is stored. Carbon can remain
in soil from hundreds to thousands of years.
4. Data from the US Department of Agriculture - Soil Conservation Service, National Soil Survey Laboratory
was used to assess what the impact of global warming would be on soil. The organic carbon of the upper
30 cm of soil was grouped by Soil Taxonomy family temperature classes. It was found that for the same
mean annual soil temperature, soils that had less than a 5°C difference in winter and summer soil
temperature at 50 cm depth ("iso") tended to have more carbon. It appears that the crucial element for
carbon storage is the maximum temperature and not the mean. Thus, to manage soil for maximum carbon
storage, extremely high soil temperatures should be avoided.
5. If soil temperatures in the temperate zone increases by 3°C, soil carbon contents would decrease by about
11%. Using this 11% value and applying it to all soils in the temperate zone, indicates that this level of soil
warming would increase the amount of atmospheric C02 by about 8%.
6. Nearly all soils could increase carbon storage if high temperatures were controlled through practices such
as mulching or shading. The wetlands throughout the world show that wet soils store high amounts of
carbon. A more cost-effective way to store carbon than creating wetlands would be to irrigate dry land,
Increased fertility gives high rates of production of organic matter. Most soils have some limiting nutrient
that could be corrected. There are extensive areas of acid soils that could have increased soil carbon if
subsoil pH was higher which would give deeper routing and greater organic matter production. Soils with
restrictive subsoil hardpans could be physically manipulated to allow deeper rooting and greater primary
production. Planting deep rooting, aluminum tolerant cultivars in acid soils would promote carbon storage
by increasing root biomass and production. Limiting tillage would increase carbon in nearly all soils because
tillage promotes aeration which speeds up organic matter decomposition.
7. Soils that have lost carbon (C depleted/degraded) and soils naturally low in fertility have the potential for
storing more carbon. An important point is that the carbon content of soils can be increased, but is society
willing to pay the costs involved? Irrigating, fertihzing, or breaking up hardpans can be expensive practices.
25
-------
William Schlesinger, Department of Botany, Duke University, Durham, NC
1. Soils are a large carbon pool; about twice the atmospheric pool and almost three times the pool in biomass;
deserts represent an additional large pool of carbon in carbonates.
2. Carbon pools: All numbers for pools will be in units of Pg of carbon (Pg = Petagrams = 1015 grams).
Terrestrial organic carbon in soil --> 1400 Pg. Of this about 50 Pg is in litter, with a low proportion
in the tropics but a high relative proportion in tundra.
Desert soils contain an additional 800 Pg, as CaCOj in the top meter of soil.
Terrestrial Biomass - > 560 Pg; Atmosphere — > 700 Pg; Oceans -- > 38,000 Pg.
3. Carbon fluxes: fluxes are in units of Pg/yr.
Fossil Fuel burning —flux to atmosphere of ca 5 Pg/yr.
Biomass net primary productivity - approx. 120 Pg/yr, about 60 Pg of this transferred annually to soil,
mostly as litter balance returned to atmosphere via respiration.
Air to ocean transfer 100-120 Pg/yr.
4. Turnover rates;
Atmosphere --> 700 Pg/(120 Pg +120 Pg) is about 3 years.
Biomass — > 560 Pg/(120 Pg) is about 5 years. (This is an average of extremes; some material lasts a
week or less while some trees are thousands of years old.
Soil -- > 1400 Pg/60 Pg roughly 20 years. Again, this is the weighted average life of carbon in the soil;
some material decomposes in hours to days whereas humins can be thousands of years old. This may
be an over-estimate of soil carbon turnover because it under-represents root turnover.
Carbonate turnover - > 3800 years. This number was arrived at by taking a known age layer in a desert
soil (Mojave desert) and dividing the carbon pool of the soil above that layer by age at the layer. This
results in an accumulation rate of 3 g carbon/m2/yr in CaC03. This rate, coupled with relevant desert
area, converts to an annualized global flux of carbon to desert carbonates of 0.023 Pg/yr.
5. Accretion of Soil Organic Carbon
Computation of the net organic accretion is more difficult than for carbonates, since there is significant
efflux of carbon from the soil after decomposition. The real question -- is there any long-term net
storage and if so, what is the rate?
Using chronosequence data (volcanic, beach or glacial retreat, etc.) to estimate, for soil systems of
known age, changes in carbon pools with soil age. Soil ages are as low as 100 years, as high as 10,000
years.
Estimated rates computed as soil pool/age; this was recognized as an imperfect approach, because in
very old systems the pool is essentially constant so the value asymptotically approaches zero. The point
is to compare young versus not-so-young soils to estimate accretion in older soils.
Key result is that young soils accumulate large amounts of carbon - 20 to 30 g/m2/yr during their first
100 or so years. From this data it appears that the carbon accumulation rate drops sharply with age,
to an average steady-state rate of about 2 g/m2/yr for 10,000 year old forest.
2fi
-------
If 2 g carbon/m2 is accumulated annually in soils of mature, unmanaged forests, that translates to about
0.40 Pg earbon/yr globally. This a soft number, intended to give a sense of the magnitude soil carbon
accretion relative to other fluxes. This value is an order of magnitude smaller than the flux from fossil
fuel burning, but well above the rate for carbonate accretion in deserts.
The rate of accumulation in soils is of about equal magnitude with the removal of carbon from
terrestrial systems as dissolved organic carbon (DOC) 0.4 Pg/yr. Therefore, additional storage of
carbon in soils appears to be diminishingly small -- in the noise. As a final note on the flux from the
soil, recent estimates of the annual accumulation of terrestrial organic carbon in ocean sediments at 0.1
or 0.15 Pg carbon/yr. It appears that much of the refractory material lost from the soil as DOC is in
fact turned over [oxidized] in the ocean before it can be sedimented.
27
-------
Richard Houghton, The Woods Hole Research Center, Woods Hole, MA
1. Recent, geophysical modeling indicates that temperate/boreal (terrestrial) regions of the earth are
accumulating carbon. If this is true, how can we find and document it? The storage of this much carbon
ought to be obvious (e.g., enlargement of tree rings) but this carbon hasn't been found and doesn't appear
to be in trees; could it be accumulating in soils? If so, how do we find and document it?
2. Documenting changes in landuse, is essentially a big bookkeeping effort-tracking all of the carbon in the
system. System carbon drops appreciably at time of disturbance/cutting/burning along with (more slowly)
oxidation of carbon in slash, wood, and soiJ over time. If there is regrowth, there is subsequent re-
accumulation of carbon in terrestrial systems with time, but if land converted to agriculture or if land simply
abandoned, there will be large long-term losses of carbon from the terrestrial system.
3. In the tropics, in 1980, deforestation was about an order of magnitude greater than reforestation. At the
present time annual net carbon release from the tropics is estimated at between 1 and 3 Pg carbon/yr. Soil
carbon losses contribute about 15-20% of total system loss; the biggest loss term is slash and stumps.
4. If the total exploitable biomass pool in the tropics is somewhere between 100 and 300 Pg, deforestation at
current levels could release up 5 Pg carbon/yr until we run out of forest. If deforestation stops immediately
and is followed by natural reforestation or by managed reforestation with controlled cutting in the future,
then carbon would be withdrawn from the atmosphere and stored in soils and biomass. The rate could be
as high as 3 Pg carbon/yr. Up to about 100 Pg of carbon could be stored. This is less total carbon storage
than was initially stored in these forests. This is because not all land is regarded as available for
reforestation and some will be supporting intensive agriculture. The actual potential for carbon sequestering
is highly uncertain - how extensive would forests become; how well would forest come back on degraded
lands?
5. Since the 1850s temperate and boreal zones have contributed significant amounts of carbon to the
atmosphere during exploitation of North America and development of the industrial era, but tropics have
dominated more recent fluxes to the atmosphere.
6. Previous data described what we "know" based on land use and related data about human use of the land.
Geophysical work presents additional data about where carbon has been coming from and going to. Glacial
ice bubbles can be used to characterize carbon content of atmosphere for past 100 to 1000+ years.
Geophysical models (ocean and atmospheric circulation and gas exchange) of how much carbon we think
the oceans are capable of removing from the atmosphere, coupled with gas bubble data, let us run models
backwards to calculate "missing" sources/sinks of carbon. Results appear to be very different for the recent
period (since 1950), land use data indicate large net release of carbon from the land while ocean models
show much smaller net release.
7. Recent modeling suggests that atmospheric carbon levels in the north temperate zone are not as high as they
"should" be, and suggest that this results from a substantial net flux into temperate/boreal terrestrial systems.
There appears to be less uptake of C02 by the oceans and more transfer to the land than previously
believed, with an unknown temperate terrestrial sink of roughly 2 Pg/yr. What does this mean? Have subtle
changes in temperature or moisture changed carbon storage? Is elevated pC02 resulting in increased
primary production? Are soils behaving differently (i.e., is carbon storage changing) from where they were
a few to several decades ago? Where/how would you look for and assess such possible changes?
28
-------
Ariel Lugo, Institute of Tropical Forestry, Rio Piedras, PR
1. One needs to put things in perspective, in terms of vegetative and soil organic carbon. Wetlands and
grasslands store the most soil carbon; tropical forests store more than boreal forests but a little less than
temperate forests in terms of overall soil carbon storage.
2. In another broad view, one can look at soil carbon in relationship to the ratio of temperature and
precipitation, which is an indicator of moisture availability. If one considers the tropics, which is more
climatically diverse than both temperate and boreal systems together, [62 Holdridge life Zones in the tropics
vs. 120 in the world], then reduce this variability to six groupings, of dry, wet, and moist for tropical,
subtropical areas: total carbon shows a nice linear decrease with increasing ratio of
temperature/precipitation (T/P) (an indicator of moisture availability) whereas relationships for soil carbon
and vegetation carbon are curvilinear, generally decreasing with higher T/P ratios, except for an increase
in vegetation carbon between T/P ratios of 0.8 to 13.
3. Within tropical forests the greatest amounts of soil carbon are in subtropical and tropical wet and rain
forests, those forests which are actually disturbed the least. If one looks at the actual data set, however, with
individual data points for soil pits with bulk density, there is lots of scatter; with climate (different T/P
ratios), soil carbon can change dramatically but within climates. There is substantial variation within life
zone associations.
4. Site topography has an impact on the accumulation of soil carbon.
5. Another factor affecting soil carbon is land use and age. For example, consider pasture sites in Costa Rica
(wet life zone) and Venezuela (moist life zone-Western grasslands). In both countries, the pastures had
greater amounts of soil carbon than did nearby mature forests.
6. Succession is another factor that affects soil carbon. Chronosequence data for the subtropical wet forest in
Puerto Rico and the subtropical moist forest on St. John, US Virgin Islands show in both instances, that the
mature forest had greater soil carbon contents than did younger forest.
7. Historical land use affects soil carbon. For example, soil carbon almost doubled, from 37-64 and from'34-60
tons/ha, respectively in moist and wet life zones, from 1950-1960 on plots abandoned following agriculture.
On the other hand, in 1980, coffee shade agriculture (a type of agriculture that uses mature trees to shade
coffee plants) had 10 tons more or almost 20 tons/ha less soil carbon, depending on whether paired plots
were in moist or wet life zones.
8. In another study following 40 years of agriculture, we found changes in organic carbon, small changes in bulk
density, and large changes in organic carbon storage (tons/ha). The annual rate of soil carbon accumulation
ranged from 20-120 g carbon/m2/yr rate. These rates are at least 10 times the soil carbon accumulation
rate of 2 g/m2/yr mentioned by Schlesinger earlier. These rates indicate that nature is very responsive to
changes in land use on a short-term basis. This means that management effects on soil carbon can be
dramatic.
9. In Costa Rica and Venezuela, using paired forest/agriculture sites, we found: 1) no relationship between soil
texture with soil carbon, 2) no difference between mature forest and agricultural sites, and 3) Venezuela soils
had 2-3 times as much soil carbon as paired Costa Rica soils. This is due primarily to the mineralogy and
high clay content of the Venezuelan soils.
10. It's important to collect and report soil bulk density data along with soil carbon data. Similarly, it's not a
good idea to infer or extrapolate bulk densities from another sites,
11. How can management play a role in increasing soil carbon storage? In answering this, I want to emphasize
the role of managed forest plantations. For example, at a site in the Luquillo Experimental Forest in Puerto
Rico, where 10 different tree species are planted in small plots, 23 yrs old, on the same soil, organic matter
percent (in top 10 cm) varied from 5.6 to 10, representing accumulations of 46 to 70 t carbon/ha. Litter
29
-------
production is a function of species.
12 Litter is not the only factor contributing to the observed differences between soil carbon in plantations and
natural forests. For paired plantation/natural forest sites we found that they tended to have similar levels
of same soil carbon in plantations and natural forest. Plantations have greatest amounts of carbon in the
litter, natural forests have greater amounts in the roots, and plantations are way ahead of natural forest in
aboveground biomass.
We then investigated the root vs. litter mass question in more detail. Over a 12-year period, a pine forest
accumulated much more litter biomass than did a paired secondary forest site. But for fine root biomass,
the relationship was inverted: secondary forest had much higher root biomass than did the paired plantation
site. When the total primary productivity was calculated, the result was the same: 19 tons/ha/yr. However,
the nature of prevalent soil inputs was very different: via litter for plantations and via roots for natural forest.
13. In summary, we need a hierarchial approach to answer the question of whether soils can be manipulated
to store more carbon. I believe that we can store addition carbon in soils via management practices to
control aboveground vegetation, such as encouraging forest succession and other things. But, consider the
task ahead. If one considers the amount of degraded lands with potential for forest replenishment, the total
is 758 million ha—the tropics have an overall total of 2 billion ha of damaged lands that have below-par levels
of carbon above- and below-ground. These damaged lands represent the opportunity we have for
manipulating succession or applying management to increase storage of carbon, both in above-ground
vegetation and below-ground.
14. Tools for storing carbon in soils:
Encourage forest succession, (i.e. greatest gains will be made in allowing/encouraging succession on
degraded sites, not in trying to improve carbon status of mature forests).
Use pastures on highly degraded lands.
Use plantations carefully, (i.e. there is great diversity on how species respond: some put more carbon
in roots, others in litter; this allows one flexibility in technique to develop different management
strategies for particular sites).
Use cultivation techniques that preserve/encourage soil structure.
Add or retain organic matter in agricultural fields, (there is a lot of "old literature" from Puerto Rico,
indicating that leaving straw in the fields from sugar cane improves soil carbon).
Recycle sewage through forest and degraded lands that have nutrient limitations.
Use plants with high root production.
Preserve wetlands and grasslands.
Manage landscapes. Think more about a complete landscape focus at watershed scales rather than
becoming entangled in the issues of hierarchies and long time scales-long-term calculations may be
correct for certain values but inaccurate in terms of (shorter scale) management objectives.
30
-------
Phil Sollins, Department of Forest Science, Oregon State University, Corvallis, OR
1. Estimates of tree root turnover are suspect and may be misleading,
2. Erosion and leaching losses of soil carbon are comparable to greater than the soil carbon accumulation rates.
We have little data on soil carbon accumulation, little for losses of soil carbon lost through erosion, and even
less for carbon losses through leaching.
A major unanswered question: Is the leaching of soil organic carbon a net sink of atmospheric carbon, or
is it mainly metabolized and degassed as C02?
3. Soil physical structure is an important factor that influences the accumulation or loss of soil carbon. Physical
structure affect soil thermal properties and water holding capacity. At the same time, soil mineralogy has
a large affect on physical structure.
4. The pools of soil carbon arc important, but the rates of change of these pools are even more important.
Based upon the compiled chronosequence data of Schlesinger, and other data, potential soil carbon loss rates
appear to be much larger than soil carbon accumulation rates. We can conclude that it's easier to lose
carbon from soils than it is to accumulate soil carbon.
5. When measuring soil carbon concentrations it is essential to obtain good estimates of soil bulk density. It's
also essential that measures of coarse fragments be made, particularly in forested soils where coarse
fragments often account for more 30% of the soil volume. Without these data, it's not possible to make
good estimates of soil carbon pools.
31
-------
John Duxbury, Department of Agronomy, Cornell University, Ithaca, NY
1. The soil organic carbon pool size is somewhere between 1400-1500 Pg. The loss of soil organic carbon due
to agriculture is between 10-100 Pg, or somewhere between 0.7 - 7% of the global soil carbon pool. There
is some uncertainty in these numbers because the data, particularly in agricultural systems are poor.
Methods generally used are poor and not standardized.
2. Ways to improve the data:
Need to sample deeper into the soil (it isn't sufficient to just sample the plow layer) - sample at least
one meter, if not more.
When reporting changes in soil carbon, the same amount of Mil needs to be compared. Thus, need to
measure and correct for bulk density variation and for stones/large fragments in the soil.
Correct for soil erosion (how much material has been physically removed?).
3. Instead of talking about various extractable soil carbon pools (e.g., humic or fulvic acids) we now talk about
soil carbon dynamics (e.g., a range of turn-over rates). I suggest four pools of soil carbon based upon
carbon dynamics.
Active or Labile Pool-Readily Oxidizable
Controlling factors include: residue inputs and the climate; type of soil is not important to this pool of
soil carbon.
Agronomic factors: cropping systems that affect residue inputs-management can affect the size of this
pool
Slowly Oxidized Pool—Macroaggrecates
Controlling factors include: soil aggregation and mineralogy.
Agronomic factors: because this pool is related to the degree of aggregation, tillage is most important-
management can affect the size of this pool
Very Slow Pool-Microapwepafes
Controlling factors include: soil aggregation and mineralogy.
Agronomic factors: it is unclear whether or not management will affect the size of this pool because it
probably involves mostly microaggregate structure which is not likely altered by tillage.
Passive. Recalcitrant Pool
This pool may not actually exist, but there is some evidence that organic compounds can become trapped
between clay plates and be unavailable or unaccessible to decomposers.
Controlling factors include: .clay mineralogy
Agronomic factors: none
32
-------
4. Evidence for and measurement of the dynamic carbon pools:
Active Pool
Evidence based upon measurements of 14C labeled residue decomposition rates.
Residue decomposition follows the same decay pattern (curve) in all environments as long as we
normalize the time scale. The rates at a given site are, however, affected by factors such as temperature
and moisture. The decomposition processes are probably the same all around the world.
The residue decomposition process appears to have two phases. The first is a rapid decay phase (about
two thirds of the carbon in the residue is lost in this initial phase) and the second phase follows first
order kinetics (Ct = C0 e~w ). Decomposition data can be used to estimate active soil carbon
accumulation rates.
The size of the active pool < 25% of the soil carbon in temperate and warmer climates, but possibly
> 25% in colder climates.
Slow Pool
Amount of soil carbon is related to the amount of aggregation.
There is no good way to measure the size of this pool, except by measuring the other pools and
calculating the difference.
There appears to be a strong positive relationship between soil organic carbon content and measures
of soil aggregates and aggregate stability,
Tilling the soil causes a shift from a system of large aggregates to a system of small aggregates. We can
destroy aggregates a lot faster than we can reform them. It is not exactly clear what causes aggregate
stability and what the role of soil flora and fauna is.
This is probably the pool from which we lose most of the organic carbon from when land use shifts from
natural systems to agriculture.
This pool ranges from 25 - 50% (or even larger) of the soil carbon.
Very Slow and Passive Pools
Evidence based upon 6 (6 = the change in 14C or 13C) content of soils when Cj plants are
replaced with C4 plants or visa versa. Cg and C4 plant type differentially discriminate in the uptake of
13C from the atmosphere. This can be used to determine how much "new" carbon has been added to
the soil. To use this technique, it's necessary to have long-term soil samples to monitor the changes in
d 13C content.
The point is that by using this 6 13C technique, evidence for the change in the level of "stable" soil
carbon is easily shown. Using this technique on a field that was shifted from Cj plants (prairie) to C4
plants (wheat) in 1888, found that is a decrease in the carbon content of the soil from 44 mg carbon/g
soil to 7 mg carbon/g soil in about 70 years.
In coarse textured soils this pool is < 25% of the soil carbon, but could be as large as 50% in fine
textured soils—where there are a lot of micro-aggregates. This is particularly true in highly weathered
soils that have lots of Fe and Al, and in volcanic soils high in allophane.
5. Soil organic matter stability is not absolute. Rather, it's a conditional parameter that's dependent upon the
actual management practices.
33
-------
Michael Beare, Institute of Ecology, University of Georgia, Athens, GA
1. Soil carbon can accumulate relatively quickly in southern Piedmont soils. If the soils are mismanaged, it can
also be quickly lost. The dispersible nature of these kaolinitic soils may help soil carbon accumulate.
2. The upper equilibrium limit for organic carbon in southern Piedmont soils is about 2.5%. The lower limit
is about 0.6%.
3. Existing soil carbon models (e.g., Meentemeyer, Post) tend to over estimate the amount of carbon found
in southern Piedmont soils. These results led to the development of conceptualization of management
practices that affect the composition and activities of soil biota that in turn affect soil carbon contents. We
hypothesized that reduced tillage agroecosystems would promote the biotic control of soil aggregate
formation and stabilization leading to more protected soil carbon and soil carbon accumulation. With
conventional tillage soil aggregates are dispersed and microbial!)' mediated organic matter decomposition
is optimized, leading to carbon losses from the soil. The extent of these effects will be influenced by climatic
and edaphic factors.
4. Conceptually, soil carbon exists in three pools: the active unprotected pool, the active protected pool, and
the passive pool. The active pool is operationally defined as that carbon associated with soil aggregation.
The active carbon pools (protected and unprotected) regulate the turnover of carbon to the slow and passive
pools.
5. Degraded soils that were studied had a lower amount of water stable aggregates than other soils. Higher
clay contents also promote aggregation, particularly macro-aggregates. Aggregates in clayey soils appear to
have less water stability than sandy soils.
6. The effect of fungal activity on aggregate stability was studied by inhibiting fungus with Captan. The percent
of water stable aggregates was less for Captan treated soils for all by the very smallest aggregate category.
34
-------
W. Mac Post, Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN
1. This discussion deals with soil carbon content as affected by temperature and moisture. This analysis is
based on data for more than 4,500 soil profiles which were corrected for bulk density and coarse fragment
content. Plotting these data on the triangular Holdridge life-zone plot shows that soil carbon content
increases as the climate is cold and wet, and decreases when the climate is hot and dry. The global average
for soil carbon content is 10 kg/m2. There is a lot of soil carbon in the moist tropics, but not much in dry
tropical climates. The tundra and tropical forests have a large portion of the total world soil carbon.
2. This approach gives a picture of carbon storage under equilibrium conditions. We're interested in how it
changes. Soil carbon content is a function of production and decomposition. The carbon turnover rate can
be calculated by;
C turnover rate = carbon in litter / turnover time.
The average carbon turnover rate of soil carbon and litter is 26 years and the rate for soil carbon is 78
years. This is not really true because there are dates of less than 100 to 1000s of years. There are
different forms of organic matter which have different turnover times.
3. To study soil carbon losses and cultivation, data for 700 paired sites were analyzed. The amount of carbon
loss depended on the initial amount. If a site started with low carbon content, the losses were low.
Conversely, if the initial carbon content was high, there was a large carbon loss. Soils lost up to 40% of
their carbon when converted to cultivation.
4. The greatest potential for increasing soil carbon through management systems is for soils that haw lost the
most carbon. Species composition of forest succession is also a factor depending the amount of soil carbon.
Aspen forest was determined to accumulate an additional 40 Mg/ha of soil carbon than pin cherry. Pastures
have higher soil carbon content than row cropping systems.
5. Soil carbon content is the result of production, species composition, nutrient decomposition, and how climate
influences all those variables. If climates change so that the temperature rises and there is adequate soil
moisture, the soil carbon content would increase, but if there was a drought carbon content would decrease.
Soil texture affects the moisture holding capacity so soil maps can be used to regionalize the effects of
changes in moisture. The timing of precipitation is an important factor to consider when modeling response
to climate change.
6. There is some concern about what is going to happen in the tundra with climate change. Boreal and tundra
soils are a net sink of 0,2 to 0.4 Pg of C02 per year. General circulation models predict that temperature
changes will be the greatest at high latitudes. If one assumes a temperature change of TC, the boreal and
tundra would contribute 1.0 to 2.5 Pg of C02 per year compared to the 5.5 Pg of C02 released annually
to the atmosphere from fossil fuel burning.
35
-------
Rattan Lai, Department of Agronomy, The Ohio State University, Columbus, OH
1. Data from ESuol and Sanchez, regarding areal distribution of soils; average organic carbon from the top 6
inches, assuming: weighted average of 1.45% organic carbon content for the tropics, 1% decline in existing
mass per year, and existing stone-free bulk density of 1.4 (maybe high) for coarse Alfisols, the total emission
was 128 billion tons of carbon per year. Conclusion: there can be a very large loss of carbon from those
soils,
2. Question: Is soil degradation somehow related to loss of soil carbon via emissions and global climate
change? Hypothesis; soil structure is related to soil carbon and if soil carbon is lost, then structure is
changed/lost. Other controlling factors of soil structure: erosion, leaching, mineralization, and
sedimentation.
3. In converting forests to arable lands, some loss of soil carbon can be attributed to surface runoff and soil
erosion.
4. Results of long-term experiments: if started with 4% organic carbon in forest ecosystem, either 100 or 25
yrs old, the amount of carbon in the top 6 inches of soil declined, regardless of which management system
was used, conventional tillage was most, no tillage a little better, and no tillage with agroforestry a Ettle
better still; yet all declined at the same rate: Similar decline results from savannahs where organic carbon
was initially 2%.
5. Is there a relationship between decay rate and soil properties? We have found that true clay content and
rainfall were the two important factors for savanna soils. Other factors: mean altitude above sea-level and
latitude (both mean greater rainfall).
6. Soil erosion data collected in watersheds for 15-18 years shows a significant negative correlation of soil
carbon loss with soil erosion: R2 value of 0.71. Data from Ivory Coast, on bow much soil carbon is lost:
forests = almost nothing, if maize on 1% slope = 2 t carbon/ha/yr, i.e. displaced some 25 m, not lost to
atmosphere. Erosional displacement selectively removes clay and organic matter only, leaving skeletal
materials behind. There is about 3-5 times more carbon in eroded material than in non eroded material.
Results from continually plowed plots on steep slopes: assuming erosional losses of 200-300 tons/ha/yr and
bulk density of 1.4, this translates to 2 cm of soil.
7. When one wants to go from a deforested system back to agricultural practices, factors to consider when
assessing greenhouse effects in term of burning are: In shifting cultivation, the farmer does not add fertilizer
(resource based agriculture versus science based). If he needs 100 kg/ha/yr of nitrogen, it can come from
the soil or from biomass additions.
8. For tropical deforestation, assuming 11 million ha of land cutover annually {range is from 4-20 million ha],
initial organic carbon content at 2%, and 10% decline in first year after clearing, then this corresponds to
40-50 million tons of soil carbon lost per year to the atmosphere.
9. Why are there big losses of soil carbon from deforestation? The important factor is increase in soil
temperature! Under forest the soil temperature is 22-23 8C. When the forest is cleared, the soil
temperature climbs to 42-47 °C, in 1, 5, and 10 cm depths. In some instances, we have monitored soil
temperatures as high as 56 or 57 °C, from 11 a.m. and 12 noon until about 4 p.m. For every 10 C increase
in temperature, the rate of chemical reactions doubles. Thus, the rate of organic matter oxidatioD increases
drastically when forest cover is removed, as both above- and belowground temperatures increase.
10. What happens if one puts mulch back on the soil surface after a forest is cleared? Under the mulch, at
application rate of 6 tons/ha, the soil temperature was lowered to 35-37 °C from 47 "C. This is still 10 °C
higher than under forest cover. Deforestation significantly changes microclimates.
11. Other mulch experiments: if no mulch is added after clearing, the organic carbon dropped from 3.3 to 1.4%
in 18 months;- if 2 tons/ha of mulch added, the decrease was still from 33 to 1.4%; at rate of 12 tons/ha
36
-------
[leaf litter addition rate in forest], the decrease was from 3.3 to 1.8%. Higher mulch addition rates were
not used. The rate of soil carbon decomposition is much higher even with mulches as compared to the
natural forest.
12. Soil profile levels of organic carbon up to 50 cm for different practices; no-till has almost 2 times the
organic carbon compared to plowed practice in the surface layer; the average at 50 cm depth is about 15-
20% greater in the no-till practice due to lower soil temperatures and greater soil moisture. One drawback-
poorer seed germination oh no-till plots, perhaps due to greater ethylene and methane production in
anaerobic (wetter) soil conditions?
Another scenario: After five years of having a degraded site from deforestation, with organic carbon
decreasing from 3 to 1%, how does one build up the soil organic carbon? Use cover crops such as legumes
and grasses. After two years, there was a 30% increase in organic carbon-results, however, were very
species dependent.
13. Influence of termites on decomposition rates? In Africa, we monitored residue decomposition from corn,
rice, and 28 other kinds of plant materials and a large proportion of them disappeared in 60 days primarily
because of termite activity.
14. Live mulches: alternate rows of corn with Stylosanthes [live mulches can out compete corn and other ag
crops for moisture in dry weather], Centrosema with corn [out competed corn also], cow peas and corn
together [worked well on steep slopes in contoured rows to prevent erosion], planting tree seedlings under
mulch in desert places of Sahel [only about 20% success rate if animal grazing can be controlled!. Leucaena
trees with coffee, fodder trees in windbreaks along fields [( Azadiracta indica) takes moisture from nearby
crop rows], alley cropping, with legume trees (Leucaenal between crop rows on steep slopes [lowers erosion
but raises organic carbon only slightly in the long-term].
15. Increased decomposition and emissions = lower soil carbon accumulation. Factors such as: mechanized
deforestation, continuous cropping, residue removal, drought-prone soils, and low-input (shifting) agriculture
that mines soil nutrients, can lead to depleted soil carbon stocks.
16. Decreased decomposition and emissions = higher soil carbon accumulation. Practices such as: maintaining
natural vegetative cover, use of cover crops, managed pastures, no-till agriculture [no-till doesn't work
everywhere], controlling soil erosion, judicious input apiculture, and agroforestry [the latter two are too
often popularized by myths rather than facts], can lead to increased soil carbon stocks.
17. How does one minimize risks to soil for losing soil carbon?
Reduce the need for deforestation by a) intensifying agriculture on existing farm lands and b) transform
traditional (shifting) agriculture into commercial agriculture-give people a reason not to move!
Minimize soil and water degradation by preventing erosion, regulating burning through effective
legislation, decrease use of marginal lands, and promote judicious use of all-farm input agriculture.
Utilize improved (good) farming systems: manual vs. machine-cleared forests; frequent use of cover
crops; conservation (perhaps no-till) tillage practices that minimize plowing; agroforestry, afforestation;
improved grafts and cultivars.
37
-------
William Parton, Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, CO
1. Using the GISS GCM to provide climate change scenarios for the Great Plains, we project that soil organic
carbon to reach new, lower equilibrium levels due to higher soil temperatures associated with global
warming. The losses are an order of magnitude less than carbon losses from cultivation, 200-300 g m'2 (over
a 50 year period) compared to M00 g m"2 (also over a 50 year period). Most of the carbon is lost from the
top 20 cm of soil. Some of soil organic carbon losses were from erosion, but we don't know how much.
In some instances, primary productivity is projected to increase because the increased temperatures promote
soil organic matter decomposition that releases organic N, which stimulates productivity.
2. Soil texture (the texture phase of the series and texture of the mineral surface horizon) was selected as an
important soil property by regressions of properties from the US Soil Conservation Service (SCS) national
soil pedon database. Soils with fine texture had higher amounts of carbon to loose when cultivated.
3. Soil texture has a large impact on the amount of carbon stabilized in the soil; fine soils have higher amounts
of soil organic carbon compared to sandy soils.
4. The CENTURY model is useful to for projecting soil fertility and organic matter. It is built around three
soil carbon pools: 1) active, 2) slow, and 3) passive. The active and slow are affected by management. The
active pool very quickly gets into equilibrium with new carbon inputs; it is manageable but is only about 3%
of the total soil carbon pool. The slow pool has about a 20 year turnover period. The passive pool is about
50% of the total soil carbon pool.
5. Using the CENTURY model to simulate different management strategies in two areas of the Great Plains:
Eastern Colorado (30 cm ppt annually) short grass prairie currently under a wheat/fallow cropping system,
and eastern Kansas (100 cm ppt annually) tall grass prairie currently under annual corn cropping system.
In the great plains, every time the soil is cultivated, soil carbon decomposition is increased by 50% for one
month
Eastern Colorado - wheat / fallow agricultural system
First 100 years of cropping - decrease soil carbon by about 40% on a fine textured soil.
How do we build back the soil carbon?
Conventional tillage plus fertilizer -- > no increase in soil carbon
No till plus fertilizer --> increase of 500 g m"2 in 100 years (5 g m^yr'1)
Grass without fertilizer --> increase of 700 g m"2 in 100 years (7 g m^yr*1)
Grass plus fertilizer:
with grazing — > increase of 1000 g m"2 in 100 years (10 g m'^yr"1)
without grazing - > increase of 1400 g m"2 in 100 years (14 g m"2^"1)
If you want to build up carbon quickly, you need to add fertilizer. Return of crop residue also
important.
Kansas - corn agricultural system
Corn with no till plus N fertilizer and returning all residue to the soil will increase soil carbon in less than
100 years to levels greater than native grassland. Native grassland had about 6800 g carbon m"2; potential
for 7800 g carbon m"2. It takes about 30 years to get from current level of 4000 g carbon m"2 to 6800 g
carbon m'2. No-till plus fertilizer out yields conventional till plus fertilizer. Residue return to the soil is
38
-------
important. The model assumes 100% return of residue with no-till and 50% return with conventional tillage.
Grazing plus fertilizer returns soil carbon levels to that of native grassland in about 50 years. Grazing
requires burning every fourth year to maintain palatability and prevent forest encroachment.
6. Forest systems stabilize the least amount of carbon in soil because of allocation of carbon in the system.
About 30% of the carbon is allocated the roots in forest system. In grassland systems about 60% of the plant
carbon is allocated to the roots. Wood carbon production shows large amounts of carbon (17000 g
carbon/m2) in 100 years. However, wood carbon may go right back into the atmosphere when the tree is
harvested. Carbon fixed in the soil will stay there for a much longer time. If you look at the short term
vs the long-term, that has a great impact on your interpretation. Wood carbon is cycled back to the
atmosphere on 15 to 30 year cycle. One needs to look at long-term carbon storage.
7. What would be the total carbon benefit possible on the Great Plains? All of the Great Plains agricultural
systems could potentially be put into no-till systems. The benefit would be that much of the soil carbon
would be recovered in a 50 to 100 year period. The dryer areas of the plains do not have the potential to
get back all of the carbon in that time frame, but the wetter sites have the potential to exceed original
amounts of soil carbon with best management options. The total amount is basically equal to the amount
that was lost. The carbon loss map also is a map of potential gain, about 1000 to 3000 g carbon/m2.
8. The amount of fertilizer used should to be determined by the amount that is needed to meet the deficiencies
of the system at a specific location.
9. Roots may have to grow deeper to acquire more moisture to support the increased growth due to N
fertilization.
39
-------
Paul E. Rassmussen, Columbia Basin Agricultural Research Center, Pendleton, OR
1. Soil erosion removes carbon. A 11,2 ton/ha erosion rate removes about 135 kg carbon/ha from a soil
profile with 20 g of organic matter/kg soil.
2. The soil is a sink for carbon. Long-term data indicate that about 18% of organic carbon added to soil in
subhumid regions is retained in the organic matter fraction (82% is respired as C02 through microbial
activity). This percentage probably increases to about 26% in cool humid agricultural regions. The
percentage is not known for forested soils or tropical soils.
3. Residue type does not have a large effect on carbon retention in soil carbon. Manure, mature legume
material, cereal straw, corn stalk and cobs, and sawdust react similarly. The amount of lignin and complex
cellulose material is a factor in retention, thus green manures may not be particularly effective.
4. Carbon input into soil is increasing in agricultural soils where erosion is minimal. The increase is promoted
by increasing the intensity of cropping (reducing frequency of fallowing) and the use of fertilizers (inorganic
or organic); both promote increased crop residue production. Grain crops (corn, wheat, sorghum) have a
greater effect than legumes (soybeans, beans, alfalfa) because a much greater amount of crop residue is
returned to the soil.
5. The amount of carbon being returned to soil is steadily increasing in grain crops concurrent with the yield
increase from improved varieties and management practices. In wheat, for example, a grain yield increase
of 85% over the past 50 years is accompanied by straw yield increase of 54%.
6. Minimum tillage practices promote carbon retention in soil. For cereal grains, retention is about 20%
greater with stubble mulch tillage than with moldboard plowing.
7. Straw burning may produce no measurable reduction in organic carbon in soil, but burning releases C02
to the atmosphere (about 65% of cereal residue is volatilized during burning), decreases microbial activity
and soil quality, and predisposes the soil surface to greater erosion.
8. Projected atmospheric C02 increases should be accompanied by increased water-use-efficiency in subhumid
areas. This should promote more vegetative growth of cool-season grasses and cereal grains. Faster
development will permit cereals to better escape summer drought-stress. Thus, unless precipitation is
drastically reduced, cereals should have greater straw production, increasing carbon-retention in soil. It is
uncertain whether the percentage retention (18-20%) would increase but it could if soil warming is not
substantial.
40
-------
Kermit Cromack, Department of Forest Sciences, Oregon State University, Corvallis, OR
1. Natural perturbations of the landscape, such as fire, set forested ecosystems back and may alter the species
composition. This has a considerable effect on the carbon content of ecosystems. Fire also affects
succession.
2. Nitrogen fixers, such as red alder, enhance the potential of forested ecosystems to accrete carbon by
increasing system productivity. Including N-fixers in silvacullural mixtures enhances the productivity of
conifer species. From an economic point of view and from a carbon storage point of view it makes sense
to use N-fixers to bolster system productivity.
3. Fire is not an unmixed blessing. Systems may take a long time to recover following an intense fire or series
of fires. Fire is a useful forest management tool that can enhance system recovery following clear cutting.
If misused, it can damage the system and slow recovery.
4. Coarse woody debris is an important carbon pool and nutrient reservoir. Microbes and pioneering tree
species use coarse woody debris. Nitrogen fixation can occur in decaying coarse woody debris. Following
fire, charcoal is an important carbon pool, with a very long residence time.
5. A large proportion of carbon is allocated belowground to support root production and maintenance, and to
support synergistic rhizosphere biology. We have little data on the carbon dynamics of these belowground
processes.
41
-------
Dale Johnson, Desert Research Institute, Reno, NV
1. Primary effects of forest management include harvesting, fertilization, and fire. Secondary effects may be
just as important and should be considered. For instance, how does carbon management affect nitrogen
dynamics? In particular, nitrogen availability, mobility, and losses through nitrate leaching?
2. Another important question is "What happens when a soil is warmed and stirred?" Carbon is lost, but what
happens to nitrogen? Eventually nitrogen is lost too, either through nitrification followed by leaching or
denitrification. If soil wanning is a consequence of climatic change we may see nitrogen being lost from
forested systems. Not only is this an important secondary effect, increased nitrate leaching may be an
indicator of soil warming.
3. Forest fertilization studies by Swedish investigators found that several years after high levels of nitrogen
(inorganic form) fertilization, none of the applied nitrogen could be found in the soil. The soil roust have
acidified and the nitrogen was nitrified to nitrate and leached. When you add nitrogen do you arrive at a
new permanent carbon increase in the system, or after a brief system response do you return to a lower site
controlled steady-state condition? The nitrogen and carbon cycles are closely related which may help explain
why we see this kind of response.
4. Fertilization with other nutrients such as phosphorous, potassium, calcium, or magnesium has a much more
longlasting effect on the site because these nutrients are less likely to be lost through leaching or
volatilization. These nutrients enhance the inorganic nutrient pool and cycle in place for a long time.
Nitrogen additions in sludge or by nitrogen-fixers may have a more permanent effect because they are
usually in organic forms.
5. There is a difference between nitrogen and non-nitrogen nutrients and their ability to increase soil carbon.
The non-nitrogen nutrients are more likely to cause more permanent increase in soil carbon, whereas,
nitrogen additions are not likely to have long-term effects on soil carbon.
6. As a management tool, controlled burns do not necessarily decrease the amount of carbon in soils.
Published data show that controlled burns results in the translocation of carbon as charcoal and particulates
into the soil profile. Fire also affects species composition, which affects belowground carbon inputs and
dynamics.
7. Early indications are that water and nitrogen limited plants will respond to elevated CO> This may indicate
increased water and nutrient use efficiencies with elevated C02. This may have an effect on litter quality
and decomposition. There may not be any mechanism for phosphorus deficient plants to improve their
phosphorus use efficiencies.
8. As the amount of wood removal increases, the C/N ratio,of the residue decreases. This leads to increased
nitrogen availability and potential decomposition.
42
-------
APPENDIX C: SUBMITTED PAPERS
Possible Mechanisms for the Accumulation and Loss of Soil Organic
Carbon in Agroecosystems on the Southern Piedmont
M. H. Beare and P.F. Hendrix
Institute of Ecology, University of Georgia
Athens, Georgia 30602
It is well known that conventional cultivation of native prairie and
forest ecosystems has contributed significantly to a depletion of soil
organic carbon (SOC) and net releases to the atmosphere (Jenny 1980,
Coleman et al. 1984, Mann 1986). The extent of these losses appears to
depend on the intensity and nature of cultivation practices, the length of
time under continuous cultivation and on soil type and climatic regimes.
Although alternative agricultural management practices (minimum or no-
tillage) have been shown to reduce SOC losses as compared to
conventional cultivation practices, it is not generally known whether
intensification of these alternative practices (increased fertilizer use,
increased primary production, return of crop residues to soils) can result
in significant accumulation of SOC on degraded soils.
The piedmont region of the southeastern U.S. extends from Virginia
into Alabama, and consists of diverse landscapes of urban, forest, pasture
and row crop land uses, Annual rainfall and temperature regimes fall
between warm, moist-temperate and subtropical. The highly weathered
Uitisols of this region contain highly dispersable kaolinitic clays that
present unique problems for SOC (and N) management. Giddens (1957)
reported rapid losses of SOC (30% in 3 years) following cultivation of
virgin forest soils on the Georgia Piedmont (Fig. 1). Other studies suggest
that after extended periods of cultivation, equilibrium carbon levels in
these soils approach 0.6-0.7%C, a decline of approximately 70% from
initial levels. Similiar amounts and rapid losses of SOC have been reported
from tropical and subtropical ecosystems under cultivation (Dalai and
Mayer 1986, Lugo et al. 1986). Accumulation of SOC also occurs rapidly in
Piedmont soils when SOC-depleted agroecosystems are converted to sod
(Fig. 1) or conservation management (Hargrove et al. 1982).
These results are in sharp contrast to observations from
agroecosystems on the North American Great Plains. First, SOC content of
undisturbed prairie soils in cool, temperate regions is typically higher
than in undisturbed thermic, udic Uitisols. Second, equilibrium levels
43
-------
after extended cultivation of prairie soils reflect total SOC losses of
approximately 30-50% (compared to 70% - Piedmont soils). Third and
perhaps most interesting, SOC toss rates appear to be much slower for
soils of north temperate ecosystems (Tiessen and Stewart 1983) than for
those of south temperate to subtropical ecosystems. Presumably, rates of
SOC accumulation in soils of north temperate ecosystems are also equally
slow.
Compared to SOC dynamics in north temperate regions, the trends in
Fig. 1 may result from basic differences in processes of SOC accumulation
and loss in warm, humid regions. We suggest that these differences are
not based solely on temperature (G10) and moisture effects, but also on
more fundamental differences In ecosystem properties. The coarse texture
(high sand and low silt content) of highly weathered Ultisols diminishes
their capacity to store SOC. Further, the kaolinitic clays dominant in these
soils are highly susceptible to slaking and dispersion (Buol 1983 etc.).
Thus, low aggregate stability and high dispersability, coupled with high
temperature and moisture regimes, promote rapid losses of SOC following
disturbance (intensive cultivation) of these soils.
Under such conditions, biotic control over the abiotic soil
environment may become more important for maintenance of ecosystem
structure and function. Central to our current research is the idea that
soil biota (esp. fungi, roots and earthworms) enhance soil aggregate
formation and stabilization in minimally disturbed soils and that soil
aggregates function in protecting SOC (and N) from rapid mineralization
(loss)(Tisda!l and Oades 1982). Our ideas revolve around the thesis that
agricultural management practices influence the composition and
activities of soil biological communities. Under reduced tillage practices,
selected soil biota may enhance soil aggreagte formation and stabilization
resulting in a larger pool of protected SOC (and N). Under intensive
cultivation, we suggest that organisms with faster turnover rates will
enhance SOC losses when protected SOC is exposed with physical
disruption of aggregates.
We have proposed the conceptual model shown in Figure 3 to explore
the dynamics of SOC in agroecosystems of the Piedmont. Although based
on other models of SOC dynamics (Parton et al. 1987, van Veen et al. 1984)
the model is intended to be operational for which all pools and flows
(fluxes) are measureable. Central to the conceptual model are the
dynamics of the active protected and active unprotected pools (after
44
-------
Elliott 1986, Gupta and Germida 1988) of SOC which may be important to
the formation of longer term and more stable SOC pools (slow/passive
pools).
Although, we do not yet have sufficient data from piedmont soils to
fit to this model, similiar relationships can be calculated for data from
the Great Plains . Table 1 shows the effects of long term cultivation {69
yr) practices on carbon pools in soil aggregates as calculated from data of
Gupta and Germida (1968). Macroaggregates declined in abundance in these
cultivated soils and resulted in a net loss of nearly 2.2 kg C m-2 as
compared to native soils. On a percentage basis, the slow/passive and
active unprotected pools of carbon remained unchanged relative to the
native soils in both macro- and microaggregates. However, the active
protected pool of carbon in macroaggregates from the native soils was
approximately 5 times greater than that of the cultivated soils.
Of the soils we are investigating from the Georgia Piedmont,
aggraded fescue soils have greater quantities of macroaggregates and
higher SOC content than those of degraded (cultivated) arable soils (Table
2). Soil texture also appears to effect soil aggregation. While finer
textured soils (Griffin sandy loam and Watkinsville sandy clay loam)
tended to maintain greater macroaggregates, these aggregates were more
susceptible to dispersion resulting in lower estimates of aggregate
stability. Soil aggregate formation and stabilization appear to be
important processes regulating SOC accumulation in these soils, however,
the specific mechanisms regulating these relationships needs further
study.
A greater understanding of the susceptibilty of different soils under
different management practices to soil carbon fluxes and of the '
mechanisms that influence SOC accumulation and loss would aid
significantly in developing soil management strategies for sequestering
soil organic carbon.
Literature Cited
Bruce, R.R., G.W. Langdale and P.F. Hendrix 1991. Soil-climate Dimensions
for crop culture technology transfer. ASA special publication (In
press).
Coleman, D.C., C.V. Cole, and E.T. Elliott 1984. Decomposition, organic
matter turnover and nutrient dynamics in agroecosystems. In:
45
-------
Agriculture Ecosystems: Unifying concepts. R. Lowrance, B.R. Stinner,
and G.J. House (eds.). John Wiley and Sons, New York. pp. 83-104.
Jenny, H. 1980 The Soil Resource; Origin and Behavior. Springer-Verlag,
New York.
Mann, L.K. 1986. Changes in soil carbon storage after cultivation. Soil
Science 142(5): 279-288.
Giddens, J. 1957. Rate of loss of carbon from Georgia soils. Soil Sci. Soc.
Proc. 21: 513-515.
Dalai, R.C. and R.J. Mayer 1986. Long-term trends in fertility of soils under
continuous cultivation and cereal cropping in southern Queensland. II.
Total organic carbon and its rate of loss from the soil profile. Aust. J.
Soil Res. 24: 281-292.
Lugo, A.E. and M.J. Sanchez 1986. Land use and organic carbon content of
some subtropical soils. Plant and Soil 96: 185-196.
Hargrove, W.L., J.T. Reid, J.T. Touchton, and R.N. Gallaher. 1982. influences
of tillage practices on the fertittiy status of an acid soil double-
cropped to wheat on soybeans. Agron. J. 74: 684-687.
Tiessen, H. and J.W.B. Stewart. 1983. Particle-size fractions and their use
in studies of soil organic matter: II. Cultivation effects on organic
matter composition in size fractions. Soil Sci. Soc. Am. J. 48: 312-
315.
Buol, S.W. 1983. Soils of the southern states and Puerto Rico. Southern
Cooperative series Bulletin 174. Raleigh, North Carolina.
Parton, W.J., D.S. Schimel, C.V. Cole and D.S. Ojima. 1987. Analysis of
factors controlling soil organic matter levels in Great Plains
Grasslands. Soil Sci Soc. Am. J. 51: 1173-1179.
van Veen, J. A., J.N. Ladd, and M.J. Frissel. 1984. Modelling C and N turnover
through the microbial biomass in soil. Plant and Soil 76: 257-274.
Elliott, E.T. 1986. Aggregate structure and carbon, nitrogen and phosphorus
in native and cultivated soils. Soil Sci. Soc. Am. J. 50: 627-633.
Gupta, V.V.S.R., and J.J. Germida. 1988. Distribution of microbial biomass
and its activity in different soil aggregate size classes as affected by
cultivation. Soil Biology and Biochemistry
46
-------
Specific Research Needs
1) Assimilation and synthesis of existing data sets from agricultural
systems to explore relationships between agricultural management
practices and SOC accumulation and loss in various agricultural
regions and soils,
2) Establish longer term experiments to evaluate the effects of changing
climatic regimes on soil carbon dynamics and carbon pools,
3) Investigate relationships between various soil C pools and management
practices on regional and soil type basis.
4) Consider mechanisms of SOC accumulation and loss (biotic and abiotic)
to advance possible management strategies for accumulating soil
organic carbon.
47
-------
Table 1. EFFECTS OF CULTIVATION ON CARBON POOLS IN SOIL AGGREGATES
(calculated from Gupta and Germlda, 1988)
Estimated C Pools (g m"2)
Soil Type/
% Water
Stable
Total C
Slow/
Active
Size Class Aggregates g m
-2
Passive Unprotected Protected
Native
>250 urn 77.1
<250 \im
22.9
5412
1614
Cultivated
>250 \im 61.4
<250 iim
38.6
3006
1853
5325
(98.4%)
1598
(99.0%)
2961
(98.5%)
1833
(99.0%)
76.4
(1.41%)
16.1
(1.00%)
42.9
(1.42%)
19.6
(1.00%)
10.9
(0.20%)
2.13
(0.07%)
-------
Table 2. Soil organic carbon and soil aggregation in aggraded and
degraded ecosystems on the Georgia Piedmont.
(M.H. Beare et al. unpublished data)
Aggregate
fraction (jim)
Aggraded Fescue Soils
HSB Griffin
Bottomland(LS) UpIand(SL)
Degraded Arable Soils
Walkinsville
(SL) (SCL)
Soil C (%)
1.60
2.33
0.73
1.00
> 2000
1000-2000
250-1 000
105-250
53-1 05
>2000
1000-2000
250-1000
1 05-250
53-105
25.80 £ 7.24
6.96 £ 0.84
6.56 £ 0.85
2.21 £ 0.27
0.95 £ 0.18
0.39 £ 0.03
0.44 £ 0,03
0.6B ± 0.02
0.77 £ 0.04
0.78 £ 0.02
% Water Stable Aggregates1
55.80 £ 5.72
6.09 £ 1.25
4.08 £ 0.66
1.17 £ 0.22
0.79 £ 0.11
17.89 £ 2.44
5.57 £ 0.26
6.30 £ 0.68
2.14 £ 0.25
1.16 £ 0.37
22.30 £ 3.07
8.54 £ 0.77
11.29 £ 1.39
3.11 £ 0.36
1.14 £ 0.14
Aggregate Stability (T2o^2)2
0.29 £ 0.03
0.38 £ 0.04
0.65 £ 0.03
0.81 £ 0.01
0.72 £ 0.02
0.30 £ 0.04
0.38 £ 0.01
0.47 £ 0.02
0.65 £ 0.03
0.74 £ 0.01
0.13 £ 0.01
0.13 £ 0.02
0.16 £ 0.02
0.30 £ 0.04
0.49 £ 0.03
1. Aggregates > 250|xm were collected from nested sieves by wet-sieving and dried @> 90° C.
Less than 250|xm aggregates were collected by gently passing the remaining suspension over a
nest of 105 and 53^m sieves. All values are corrected for primary particles by size class.
Values = x + 1 S.E., n=4.
2. Based on a turbidimetric analysis after Williams et. al. (1966). Aggregates were seperated
by wet-sieving and air-dried on the sieves. Intact, pre-wetted aggregates (0.25g d.w.) were
placed on an end-over-end shaker for 2 and 20 mtns, let settle for 30 mins, and transmittance
measured (@520nm) with a spectrophotometer (Spec 20). ValueseTgQ/T^; x £1 S.E., n-4.
LS - Loamy sand
SL - Sandy loam
SCL - Sandy clay loam
49
-------
Fescue
Kudzu
Forest
GIDDENS (1957)
JONES etal (1966)
Fescue
Alfalfa
0 20 40 60 80
YEAR
Figure 1. Chronosequences on the Georgia Piedmont show rapid losses
of soil organic carbon when undisturbed ecosystems are
plowed, and rapid accumulation when intensive cultivation
of agroecosystems is ceased. Equilibrium levels appear
to range between 0.6% C and 2.3% C. (R.R. Bruce et al. 1991)
-------
A
PLANT
PARTIOUI ATE
I # 11 1 I I V»/ W 1 !¦»
-~(harvest
coe
FAUNA
FUNGI
XFKXJT1S
ACTIVE
UNPROTECTED
S*
(W)
SOIL
SOLUTION
FOOTS
FUNGI
MACROFAUNA
FUNGI
ROOTS
ACTIVE
PROTECTED
¦FUNGI
BACTERIA
FUNGI ~
BACTERIA
mss
LEACHING
EROSION
PASSIVE
Figure 3. Conceptual model of carbon dynamics. Boxes represent
Unaa
-------
Pedogenesis of Carbon 1n Soils
S, W. Buol
Two major forms of carbon are present in soils. Carbonates, especially
CaC03, are abundant in arid parts of the world where the carbonate is inherited
from geologic materials and often distributed via dust. Pedogenically C02
is released from carbonates if the soil becomes wetter or more acidic. It
is usually translocated downward and precipitated in subsurface horizons of
any soil receiving even slight precipitation.
CaC03 + H20 + C0£ v* Ca(HC03)2
insoluble soluble
Organic carbon results from the decomposition of plant, animal and
microbial biomass both on the soil surface or near surface as in the case
of soil microbes and plant roots.
The soil is but a brief repository of carbon in its cycle. The following
schematic illustrates pedogenic pathways available for organic carbon:
Soil Organic Carbon
Dissolved ]
carbon in v 4— leached
groundwater I
humifi cati
chai
The amount of organic carbon in a soil represents a steady state between
rate of organic carbon production, incorporation and oxidation. Production
oxidized
C0„ + H
52
-------
rates can be estimated from above and below ground biomass growth rates.
Oxidation rates are subject to control by soil temperature, supply in the
soil, and nature of the multitude of humus forms along the humification chain,
Soil organic carbon contents related to soil temperature are represented
in this figure of data in the National Soil Survey Data base. The higher SOC
contents in iso-soi1 temperature regimes probably reflect lack of summer
extremes in the tropics versus the temperate zone.
Within any given climate, those parts of the landscape where the soil
is saturated with water for long periods of time, excluding 0^ from the soil,
have higher organic carbon contents than do better aerated soils. The extreme
of this condition results in oxidation rates slower than biomass production
and Histosols are formed. Subsequent burial of such soils has formed our coal
deposits.
How long carbon is retained in the soil varies greatly. Age of soil organic
matter determined by radio carbon dating indicates deeper samples are older
than near surface samples. Within A1fi sols and Mollisols, age values extend
from 200 year B.P. at the surface to 7000 years B.P. for samples at 150 cm.
Spodosols, notable for their release of organic carbon to surrounding ground- -
water, ultimately forming "black water rivers," have carbon ranging in age
from a few years to only about 2000 years B.P.
53
-------
30
Q,
to
•o
E cC"
o •
CO
Q
O
O)
20
10
\
Non-iso
4,5
11.5
18.5
25.5
Family Mean Temperature (°C)
54
-------
At a United Nations Environmental Protection (UNEP) meeting in Nairobi
two weeks ago a subcommittee formulated the following outline of conditions
that can be expected to increase the content of carbon in soil, albeit a new
steady state will be reached in response to the altered condition.
Soil Change
Cooler soil
Wetter soil
Increase fertility
Increase subsoil pH
Fracture subsoil pans
Deeper rooting
Reduced aeration
Practice
Mulch, shade
Irrigation
Mineral fertilizer
Deep liming
Subsoiling
A1 tolerant cultivars
Limited tillage
Applicable Soi1s
All soils
Dry areas
Most soils
Acid subsoil
Hardpan soils
Acid subsoils
All soils
55
-------
MANAGING SOU- CARBON IN TROPICAL AGRO-ECOSYSTEMS
R. Lai
Department of Agronomy, The Ohio State University
Columbus, Ohio 43210
Summary
World soils have a potential to immobilize atmospheric carbon as soil organic
matter and reverse the trend of increase in atmospheric concentration of carbon dioxide.
Globally, world soils contain more carbon than atmosphere or worlds biota. Misused,
soils can be a major source of carbon emission into the atmosphere. Agricultural
operations that enhance emission of carbon into the atmosphere as C02, CH4 or CO include
deforestation, burning, intensive cultivation, and manuring. Soil degradation by
accelerated erosion, compaction, leaching and acidification also cause depletion of
soil organic matter and emission of carbon into the atmosphere. In contrast, soils can
be a major sink for carbon through judicious land use and proper soil management.
Restoration of degraded lands through afforestation and management of these man-made
forests as carbon sinks would fix carbon back into the soil. Soil-enhancing
agricultural practices include conservation tillage, mulch farming, agroforestry,
judicious use of off-farm input, transformation of resource based subsistence farming
into science-based and market-oriented agriculture. Sustainability of agricultural
systems should be judged in terms of their effect on global carbon.
56
-------
Introduction
The "greenhouse effect" is a popular term denoting the wanning of atmosphere due
to attenuation of radiation by elevated levels of radiatively active gases in the
atmosphere. Technically, greenhouse effect is the difference between the planetary
surface temperature (Ts) and a radiative temperature (Tr). In the absence of the
atmosphere, Tf equals T(. Other than water vapors, most important radiatively active
gases in atmosphere include derivative of carbon such as C02, CO and CH4. These and other
trace gases permit the short-wave radiation reach the earth surface relatively
unattenuated. However, the main effect comprises of absorption by these gases of the
long-wave outgoing radiation and re-radiating some of it back to earth. Consequently,
the mean global temperature is believed to have been increased by 0.5 - 0.7°C over the
last 100 years. If the present trend continues, the global average temperature may
increase by 2 to 6°C during the next century (Schneider, 1989).
World's soil contain more organic carbon as soil organic matter than world biota
or the atmosphere. The size of soil carbon reservoir is about twice that of the
atmosphere (Stevenson, 1982; Sedjo and Solomon, 1989). In fact, soil organic matter is
a major active reservoir in the global carbon cycle. However, there are few reliable
estimates of its size and rate of turnover (Moore et al., 1988). The lack of knowledge
about this major reservoir of carbon is due to several factors. We have reliable
estimates of different soil types and major soil groups, estimates of the amount and
temporal variations in organic matter content for different soils are not known.
Reliable estimates of the areas devoted to different land uses and cropping/farming
systems are also not available. Furthermore, it is difficult to obtain an accurate
record of rate of change of land use systems on a global scale. Perhaps the most
important missing link in assessing contributions of soils to global carbon budget is
the lack of information about response of carbon reservoir in soil to changes in land
use.
57
-------
There has been a steady increase in atmospheric concentration of C02 and other
greenhouse gases into the atmosphere since about 1850. The increase in atmospheric
concentration of C02 is attributed to several factors, e.g. deforestation and release
of C02 from the biomass by decomposition or burning (Houghton, 1987), and combustion of
fossil fuel (latch 1986). Methane is produced in swamps and flooded rice fields
(Mikkelsen, 1988). However, the soil as a potential source or sink for atmospheric
concentration of COz, CO, or CH4 has not been given the attention it deserves. Tans et
al. (1990) observed that oceans alone do not account for the possible sink of carbon.
The magnitude of the unknown sink is about 2 to 3.4 Gt of C per year. The mechanism of
this C sink is unknown, and may be related to terrestrial ecosystems.
The objective of this report is to highlight the importance of soil as a source or
sink for atmospheric carbon, and to discuss soil and crop management, and land use
systems that may mitigate or enhance the warming trend. Preliminary calculations are
presented to show that world soils can, in fact, be a major factor in global carbon
balance. Specific examples are cited from some soils of the tropics.
Dynamics of Organic Matter in Soils
Carbon in soil organic matter, as a dynamic entity, is a function of numerous
interacting factors that include steady addition by biota and agricultural operations,
and depletion by biochemical changes, leaching and soil erosion (Figs. 1 and 2). A
simple model to predict the rate of change of organic carbon in soil proposed by several
researchers (Greenland and Nye, 1959; Greenland, 1971; Jenkinson and Raynor, 1977;
Stevenson, 1982) is shown in Equation 1:
¦ ~KC + a (Eq. I)
where K is the decomposition constant, C is the carbon content of a given mass of soil
at time t, and a is the accretion constant giving the amount of carbon added to the given
mass of soil in unit time through biomass, agricultural operations, etc. It is the
difference between KC and a that determines whether soil serves as a source or sink for
58
-------
C. Soil degradation by different processes would lead to a net emission of carbon from
soil into the atmosphere (Fig. 2). Soil becomes a net contributor of carbon to the
atmosphere when KC exceeds a, but serves as sink when KC is less than a. Depending upon
soil and crop management and ecological environment, soil attains a mean value of
organic carbon CB. In that case. It is possible to consider the factor KCk that
determines whether soil is source or sink for atmospheric carbon. Agricultural
practices that affect the magnitude of K are listed in Table 1.
There is a similarity between mineralization of N and C in soil. With that
assumption, the following analysis for C is adopted from that proposed by Greenland
(1971) for N. The amount of carbon in soil depends on management, e.g. duration of
cultivation vs. fallowing. Lengths of cropping (tc) and fallowing (tf) can be adjusted
to attain the desired level of carbon in soil. The emission of carbon during cropping
phase is ctc during the fallowing phase is ftf * mean amount of
organic carbon during these periods is approximately Cr When steady state is attained
the amount of carbon emission from soil must equal the amount stored into the soil
according to the law of conservation of mass (Equation 2).
(-KcCm + ac) tc + (-KfCm + af) tf - 0 (Eq. 2)
re-arrangement of Equation 2 leads to Equation 3 that can be used as a guide to attain the
desired ratio of cultivation to fallow phases for soil enhancement.
, af ' Kfcm (Eq. 3)
'f KeC„ " ae
If K and a are known experimentally, CB can be calculated for a given land use system.
Equation 3 can be used to develop a national or a regional soil policy regarding the land
use intensity or land use factor.
If soil is subjected to continuous cultivation, carbon content of soil declines
exponentially until an equilibrium (Ce) value is attained. The magnitude of C# depends
on cropping system, soil type, and the climatic regime (Fig. 3). The difference (AC) in
initial (Ce) and the equilibrium level of carbon (C#) is approximately that emitted into
59
-------
the atmosphere or carried in water as dissolved carbon. The amount of carbon in soil at
time t is given by Equation 4:
C-C. + (C„ - Cg) e"rt (Eq. 4)
where C is soil organic carbon at a time t, r is fraction of C decomposed peT year and t
is time in years. The magnitude of C is a function of management and environmental
characteristics.
Soil as Source of C
The magnitude of decomposition constant K is generally more for tropical than
temperate environment. The Van't HofFs law states that the rate of any chemical
reaction approximately doubles for every 10°C rise in temperature. All other factors
remaining the same, the rate of decomposition is expectedly more in the tropics because
mean soil temperature is higher than in temperate regions. The amount of carbon in soil
is, therefore, related to temperature (Equation 5).
C=ae"ICT (Eq. 5)
where T is mean annual temperature.
The amount of carbon that can be emitted from soils of the tropics can be computed
from the data in Table 2. The weighted mean carbon content of the soil computed from the
data in Table ] is 1.45%. Assuming the bulk density of top 15-cm layer to be 1.4 Mg m"3
and the rate of decrease of carbon due to cropping to be 1% per year, the amount of
emission from soils of the tropics equals 30 - 45 Mg C/ha/yr or 127.9 x 10' Mg C/yr. This
is a large amount, indeed. In addition, new land is annually being brought under
cultivation. The rate of new land development is- approximately 11 million hectares
annually (Lai, 1987). The initial soil organic carbon in the top 15-cm layer is about
2%. The rate of loss of carbon due to cultivation in the first year may be as much as 10%
(Lai, 1981). Assuming the bulk density value of 1.4 Mg m*3, carbon emission from newly
cleared land is 46 x 106 Mg/yr,
There are several soil-related processes that accelerate the rate of carbon
emission from soil. These include respiration and exposure of organic matter to roicro-
60
-------
organisms, dissolved carbon in drainage water ud overland flow, and entrainmeot of
carbon with eroded sediment. Because of a preferential removal of organic matter,
accelerated soil erosion rapidly depletes carbon content of the soil. Soil carbon
content is, therefore, negatively correlated with the amount or severity of past erosion
(Lai, I980)(Equation 6),
Carbon content of soil (%) - 1.79 - 0.002E, r - -0.71 (Eq. 6)
where E is soil erosion in t/ha/yi. Humus is an important component of stable
aggregates. Organo-mineral complexes are blocked within micro-aggregates as binding
agents. In stable aggregates, resistant to detachment by raindrop impact, these organo-
mineral complexes are even inaccessible to micro-organisms. Soil erosion also removes
the clay fraction, from soil. Clays stabilize organic matter content in the soil. There
exists a positive correlation between carbon and clay contents of a soil (Stevenson,
1982).
Agricultural Operations and Carbon Emission
There are several agricultural practices that enhance carbon emission from soil
(Table 3). Conversion of tropical rainforest and burning directly impact carbon release
from the biomass. In this regard, shifting cultivation and related bush fallow systems
plan an important role. When shifting cultivation is practiced with a long fallow phase
and a high Land Use Factor1 (L >10), carbon balance is favorably maintained with
relatively high storage in soil and the biota. However, increase in the duration of
cultivation phase and drastic reduction in that of the fallow phase leads to soil
degradation and emission of carbon into the atmosphere. In contrast to deforestation
and intensive cropping, agriculture practices that would increase carbon storage in
soil include afforestation, pasture establishment and cover crops with controlled
grazing and low stocking rate. Resource-based agricultural practices that mine soil
fertility would enhance carbon emission from soil. For example, to harvest an
1 Land Use Factor L = t£ + tf/tc
61
-------
equivalent of 100 Kg N/ha in b crop would require mineralization of 1000 Kg of carbon
from humus considering a C:N ratio of 10:1. If used as low-input resource-based
agriculture, the total cultivated area of 5 * 109 ha in the tropics would emit 5 x 109 Mg
C/yr. On the other hand, a large proportion of this carbon can be retained in the soil
if fertilizers or other amendments were used as is the case in science-based
agriculture. In addition to socio-economic and political considerations, there is a
strong environmental justification for transforming resource-based subsistence farming
into science-based and market-oriented commercial agriculture.
In addition to agro-economic considerations, sustainability of agricultural
practices must be evaluated in terms of their effectiveness in minimizing risks of
carbon emission from soil. Some of those practices that have potential to reduce carbon
emission from soil-related processes are listed in Table 4. There are three principal
considerations. Firstly, it is important that the need for bringing new land under
cultivation is reduced by intensively farming existing land and by transforming
subsistence and extensive farming into intensive agriculture. Secondly, resource
management policy must be adopted to minimize soil and environmental degradation.
Degradative trends can be reversed by preventing or decreasing soil erosion, regulating
burning and grazing, decreasing use of steep/marginal lands for agriculture, and
through a judicious use of off-farm input. Judicious use of organic soils is an
important consideration. Globally there are 240 million ha of peat soils (Table 5).
Once cleared and drained, these soils are highly susceptible to degradation. The rate
of decomposition of carbon from organic soils is far greater than those of mineral soils.
It is not uncommon to lose 2 m of organic soils in less than a decade. Thirdly, adoption
of improved best-management-technologies must be vigorously pursued. These
technologies are often site-specific and have to be locally validated and adapted. Use
of improved crops and cultivars, conservation tillage, agroforestry and planted fallow
can drastically reduce carbon emission from soil.
62
-------
Soil As a Carbon Sink
Soil, the . uppermost layer of earth that is agriculturally productive, is a major
component of potential terrestrial sink that can be used for carbon sequestering. A
viable alternative to exploit this vast sink for reversing the trend in global carbon
emission is to restore degraded soils. It is estimated that for l.S billion ha of
currently cropped land, an additional 2.0 billion ha of once biologically productive
land has been rendered unproductive through irreversible degradation (UNEP, 1986).
Some surveys have estimated that soil degradation, of one type or another, affects about
one-third of earth's land surface. The cunent rate of soil degradation is estimated to
be 5-7 million ha per year with a potential to increase to 10 million ha per year by the
year 2000 (FAO, 1983). From a global perspective, the first priority should be to
restore biological productivity of these soils and use them as a terrestrial carbon
sink. Afforestation for using these lands to produce biomass, and intentionally
maintain them as a sink of carbon, would immobilize atmospheric carbon into biota, and
drastically increase carbon flux into the soil. Such carbon-sinks should be financed
by international agencies.
In addition, soil-enhancing agricultural practices must be adopted for arable
land uses. Use of crop residue mulches (Table 6), conservation tillage (Table 7),
agroforestry system (Fig. 4), cover crops and planted fallows (Table 8) are proven
technologies that minimize risks of soil degradation and maintain favorable level of
carbon in the soil. There is a need to develop regional, national and international soil
policy to adopt these soil-enhancing practices. Farmers should be given incentives and
encouragement to adopt those cultural practices that encourage influx of carbon into the
soil.
Conclusions
World soils have a potential to mitigate the global warming risk through their
capacity to immobilize carbon as soil organic matter or humus in the root zone.
'63
-------
Improperly used, soils are also a major source of carbon emission into the atmosphere.
Carbon emission from soil-related processes is enhanced by bio-degradation of soil
organic matter. The latter depends on temperature, moisture regime, cultivation
systems and managerial input. Depletion of soil organic matter and its release into the
atmosphere is enhanced by deforestation and biomass burning, plowing, intensive use of
marginal lands and reduction in the length of fallow phase. In contrast, carbon influx
into the soil can be increased by restoring productivity of degraded lands through
afforestation and maintaining these man-made forests as carbon sinks. Soil-enhancing
agricultural practices, e.g. agroforestry, conservation tillage, mulch farming, etc.
should be encouraged through proper incentives. There is a need to develop regional,
national and international soil policy toward using world soils as sink for atmospheric
carbon.
64
-------
References
Batch, W. 1986. The gases and their influence on climate. Natural Resources and
Development 24:90-124.
Beek, K. J., W. A. Blokhuis, P. M. Driessen, N. Van Breeman, R. Brinkman and L. J. Pons.
1980. Problem soils: their reclamation and management. International Soil
Reference and Information Centre, Wageningen, The Netherlands.
FAO/UNEP. 1983. Guidelines for control of soil degradation. Land & Water Div., FAO,
Rome, Italy.
Greenland, D. I. and P. H. Nye. 1959. J. Soil Sci. 10:284-299.
Greenland, D. J. 1971. Changes in the nitrogen status and physical condition of soils
under pastures, with special reference to the maintenance of the fertility of
Australian soils used for growing wheat. Soils & Fertilizer Abstract 34(3):237-
251.
Houghton, R. A. 1987. Terrestrial metabolism and atmospheric C02 concentrations:
independent geophysical and ecological estimates of seasonal carbon flux address
global change. Bio Science 37:672-678.
lenkinson, D. S. and J. H. Ray nor. 1977. The turnover of soil organic matter in some of
the Rothamsted classical experiments. Soil Science 123:298-305.
Juo, A. S. R. and R. Lai. 1977. The effect of fallow and continuous cultivation on* the
chemical and physical properties of an Alfisol in the tropics. Plant Soil 47:567-
584.
Juo, A. S. R. and R. Lai. 1979. Nutrient profile in a tropical Alfisol under
conventional and no-till system. Soil Sci. 127:168-173.
Lai, R. 1987. Conversion of tropical rainforest agronomic potential and ecological
consequences. Adv. Agron. 39:173-264.
Lai, R„ G. F. Wilson and B. N. Okigbo. 1979. Changes in properties of an Alfisol by
various cover crops. Soil Sci. 127:377-382.
65
-------
Lai, R., D. De Vleeschauwer and R. M. Nganje. 1980. Changes in properties of a newly
cleared tropical Alfisol as affected by mulching. Soil Sci. Soc. Am. J, 44:827-
837.
Lai, R. 1981. Clearing a tropical forest. D. Effects on crop performance. Field Crops
Res. 4:345-354.
Lai, R. 1989. Conservation tillage for sustainable agriculture. Adv. Agron. 42:85-
197.
Lai, R. 1989. Agroforestry systems and soil surface management of a tropical Alfisol.
m. Soil chemical properties. Agroforestry Systems 8:113-132.
Mikkelsen, D. S. 1988. Technological options for reducing emissions. U.S. - EPA
Workshop on agriculture and climate change, March 1988, Washington, DC.
Moore, B,, M. Gildea, C, Verosmarty, D. Skole, J. Medillo, B. Peterson, E. Rasteller and
P. Steudder. 1988. Biochemical cycles in global ecology; Towards a science of
the biosphere. M. B. Ramble et al. (eds.) Academic Press, New Yorkil 13-141.
Sedjo, R. A. and A. M. Solomon. 1989, Climate and Forests. In: Rosenberg, N. J. et al.
(eds.) "Greenhouse Warming: Abatement and Adaptation". Resources for the Future,
Inc.,-Washington, DC.
Stevenson, F. J. 1982. Humus chemistry, genesis, composition reactions. J. Wiley &
Sons, New York, 443 pp.
Tans, P., I. Y. Fung and T. Takahashi. 1990. Observational constraints on the global
atmosphere C02 budget. Science 247:1431-1438.
UNEP. 1986. Farming systems principles for improved food production and the control of
soil degradation in arid and semi-arid tropics. ICRISAT, Hyderabad, India.
66
-------
Table 1. Effects of agricultural operations on relative magnitude of the decomposition constant
High K Low K
Mechanized Deforestation Natural Vegetation Cover
Plowing Planted Fallows
Continuous Cropping Afforestation
Accelerated Erosion Cover Crops
Drought-Prone Soils Managed Pasture
Residue Removal No-Till
Low Input Agriculture Erosion Control
Soil Fertility Depletion Judicious Input
Agroforestrv
Tabic 2. Average organic carton contents and hectarage of some tropical soils (0-15 cm).
Area Organic Carton
Soil (106 ba) <%)
Oxisols 1,100 2.07
Ultisols 550 1.39
Alfisols 800 1.30
Aridisols 900 0.75
Entisols 400 1.50
Inceptisols 400 1.50
Mollisols 50 2.44
Lai (1986), Sanchez (1976), Greenland et al. (1989)
67
-------
Table 3. Agriculture] practices in the tropics thai enhance the greenhouse effect
• Burning
• Deforestation
• Intensive and Continuous Upland Farming
• Rice Paddies
• Pasture
• Chemical Fertilizers
Table 4. Agricultural practices that mimmiTf carbon emission from soil.
I. Reducing Need far Deforestation
a. Intensive Farming cm Existing Land
b. Transforming Traditional Into Commercial Agriculture
II. Minimizing Soil and Environmental Degradation
a. Prevent Erosion
b. Regulate Burning
c. Decrease Use of Marginal lands
d. Judicious Use of Off-Farm Input
e. Judicious Management of Peat, Muck and Other Organic Soils
HI. Adopt Improved Farming Systems
a.. Manual/Shear Blade Gearing
b. Frequent Use erf Cover Crops and Planted Fallow
c. Conservation Tillage
d. Agroforestry
e. Afforestation
f. Improved Crops and Cultivars ''
68
-------
Table 5.
Global distribution of peat soils {Beek et al., 1980).
Regioo Area
(ID4 ha)
Africa 12.2
Near and Middle East -CI-
As ia and Far East 23.5
Latin America 7,4
Australia 4.1
North America 117.8
Europe 75.0
World Total 240.0
Tabled. Mulching effects on soil organic C (Lai et aL, 1980).
Mulch Rale C at Different Times (Months') After Clearing
T/Ha/Yr 0 12 18
0 3.3 1.7(52%) 1.4(42%)
2 3.2 2.0 1.4
4 3.2 2.0 1.5
6 3.2 . 2.3 1.7
12 ' 3.2 2.5 1.8
69
-------
Table 7. Organic carton profile in a no-till system (Juo and Lai, 1979).
Organic Carton <%)
Depth (cm) No-Till Plowed
0-5 2.87 1.17
5-10 1.77 1.19
15-15 1.23 1.10
15-20 1.11 1.09
20-25 0.71 0.91
25-30 0.96 0.82
30-35 0.57 0.82
35-40 . 0.47 0.62
40-45 0.42 0.47
45-50 0.34 0.37
I 1.05 0.86
Table 8. An example of carbon sequestering in a tropical soil by planted fallows and cover
cropping (Lai et al, 1979).
Soil organic carbon content (%)
Cover crop Initial in 1974 Final in 1976 4b Increase
Brachiaria
1.21
1.57
29.8
Paspalum
1.23
1.45
17.9
Cynodon
1.30
1.70
15.4
Pueraria
1.27
1.50
l&.l
Stylosanthes
1.30
1.63
25.4
Stizolobium
1.30
1.57
20.8
Prophocarpus
1J0
1.57
30.8
Cenrrosima
1.30
1.53
15.4
Weed fallow
1.33
1.37
3-1
LSD (.05)
0J0
0.23
70
-------
Organic
Amendments
Carbon
Addition
In Soil
(Humltlcatlon)
Blomass
Carbon
Root Mass
Fig. 1. Carbon addition to the reservoir of soil organic matter.
71
-------
Mineralization
of
Soli Organic
Matter
Leaching
and
Acidification
Carbon
Emission
From Soli
( Soil Structure
Deterioration
Soil Erosion
and
Sedimentation
Fig. 2. Processes of soil degradation leading to carbon emission from soil.
72
-------
£
s
M
c
m
QJ
O
2 -
Time (yrs)
NT + CC + AF
Fig. 3. A generalized curve showing changes In soil organic carbon with lime
after deforestation of tropical rainforest and conversion to arable
landuse. CT: plow-based farming, NT: no-til! farming, CC: cover
crop, and AF: agroforestry.
73
-------
Plow-ti
No-til
0-5 cm
o—-o 5-10 cm
\
Leucaena -4m
Leucaena -2m
Gliricidia - 4 m
Gliricidia -2m
LSD (0,1)
Systems (S) 0.12"
Treatment (T) 0.15
Depth (D) 0.12-
Year (Y)
Fig. 4
3 4 0 1
Years of cultivation
Changes in soil organic carbon content in agroforestry
versus crop-based systems (Lai, 1989)
74
-------
Lugo, A.E., and S. Brown, Management of Tropical Forest Lands for Maximum Soil Carbon Storage
INTRODUCTION
Our research has focused on the accumulation of soil organic carbon (SOC) in tropical forests under various
types of land uses. Results of our studies are summarized below according to major land use. We then discuss
some general aspects of our work and suggest management strategies for maximizing SOC in tropical forest
lands. Finally, we discuss research needs.
RESULTS
Mature Forests
The Holdridge life zone system provides an objective framework for studying large scale patterns of SOC content
and distribution. Life zones embody the effects of climatic factors on ecosystem structure and function. We
found that, in mature tropical forests, SOC increases with available moisture.
However, to fully understand the dynamics of SOC, it is necessary to study smaller spatial scales. Within plant
associations (the next lower level of the life zone system of Holdridge), SOC may be different. The factors that
regulate SOC accumulation at this scale are biotic, edaphic, and topographic factors. At the third level of the
hierarchical life zone system, i.e., the successional stage, time and land-use factors come into play. The age of
a stand, its land-use history, and current use status, all have a measurable effect on SOC.
After Forest Conversion
Soil carbon decreases immediately after conversion of mature forests to other uses. However, the degree of the
reduction varies according to the type of conversion, the intensity of use, and the length of time a soil is under
a given use status.
Agricultural Systems
Agricultural systems are usually characterized by low amounts of SOC. In comparison to adjacent mature
forests, agricultural systems have lower amounts of SOC. Intensive agricultural use with low regard to
fertilization or organic matter management reduces SOC. However, there appears to be a minimum SOC below
which no further loss occurs. Addition of straw or organic debris to agricultural soils helps improve their SOC
content and their nitrogen fertility.
Pastures
Pastures accumulate high amounts of SOC. We have measured as much or more SOC in pastures as in adjacent
mature forests. Improved pastures have higher SOC than non-improved ones. High root production by grasses
may explain why pastures accumulate so much SOC.
Tree Plantations
Tree plantations accumulate SOC as they mature. The rate of SOC accumulation in tree plantations is species-
dependent. Intensively-managed plantations can accumulate more litter and do so faster than unman aged
plantations or natural successions. The accumulation of litter in plantations is species dependent. The
plantations that we have studied exhibit nutrient and organic matter dynamics that result in high magnitude of
accumulation in various ecosystem compartments (e.g., vegetation, litter, and soils). In contrast, we have
observed that natural successions have dynamics that result in lower accumulations of nutrients and organic
matter, but faster turnovers.
75
-------
Secondary Forests
In spite of the fast turnover of nutrients and organic matter, tropical secondary forests accumulate SOC as they
mature. The rate of accumulation of SOC is dependent on available moisture and previous land use.
Comparisons of organic matter budgets of plantations and natural forests suggest that different compartments
in these ecosystems have different behaviors. The accumulation of SOC is through different pathways. For
example, root productivity appears to be higher in natural forests as opposed to plantations. But plantations tend
to accumulate more litter.
General Aspects
The life zone condition (climate) is a good predictor for SOC accumulation and behavior for large scale analyses.
Soil texture helps explain the magnitude of SOC, particularly the sand fraction. Sandy soils tend to accumulate
less SOC than clay soils. However, our results from Costa Rica and Venezuela do not support the sand-SOC
relationship. Root biomass and turnover may be more important determinants of the accumulation of SOC in
tropical forests than litter input, at least in the short term.
Management Implications
Soil organic carbon is a long-term carbon sink with a slow turnover. As such it is a more secure carbon
sequestering mechanism than plant biomass. However, the accumulation of SOC need not be slow. Our results
show fast rates of accumulation in successional forests, tree plantations, and pastures. An obvious management
recommendation is that by allowing the growth of successional systems, SOC accumulation is also promoted.
Because of the apparent relative importance of root production and turnover to SOC accumulation, the
establishment of systems with high rates of these processes should be considered.
A critical question is the recovery of SOC in degraded or damaged lands. Land rehabilitation efforts can begin
with the establishment of grasses and pastures. These plants modify site conditions, favor future tree
establishment, and immediately favor SOC accumulation.
Tree plantations using exotic species offer another mechanism for the rehabilitation of damaged lands and for
increasing SOC accumulation. Species selection is a critical factor for consideration when artificial systems are
used to maximize SOC accumulation. Some species have high litter productivity but low root production. Others
have the opposite pattern. And, each species produces organic litter (roots, wood, and leaves) of different
qualities and suitability for becoming SOC. We have documented that at least for two plantation species, pine
and mahogany, the rate of root production is lower, and the rate of litter production higher, than comparable
secondary forests.
Acceleration of SOC production can also be accomplished by accelerating succession or ecosystem productivity.
This can be accomplished by fertilization and watering. However, both of these actions are expensive. Yet,
where a supply of domestic sewage is available, it is possible to couple land rehabilitation with the application
of treated sewage. A side benefit of this strategy is sequestering SOC.
The most fundamental approach for the management of SOC on a large-scale basis, is to implement a system
of landscape management. Such a system management takes advantage of the natural patterns of SOC
accumulation and ecosystem productivity. An objective of such management should be to maximize organic
matter productivity and ecosystem values, while optimizing SOC conservation and accumulation.
Research Needs
More attention on total carbon budgets is needed. Studies must be constrained by: life zone, edaphic and
topographical factors, age of stands, and land use (past and present). More attention on SOC turnover and
quality (they are related) is also a priority. For a general understanding of this problem we will need
standardized methodology. The Tropical Soil Biology and Fertility Program of IUBS and MAB programs is a
good example to follow. We also need to develop and test landscape-level management schemes that focus on
the productivity and biodiversity of the land and seek long-term sustainable solutions to the question of natural
76
-------
values.
Literature Consulted
Brown, S. and A.E. Lugo. 1982. The storage and production of organic matter in tropical forests and their role
in the global carbon cycle. Biotropiea 14:161-167.
Brown, S. and A.E. Lugo. 1990. Effects of forest clearing and succession on the carbon and nitrogen content
of soils in Puerto Rico and US Virgin Islands. Plant and Soil (in press).
Brown, S. and A.E. Lugo. 1990. Tropical secondary forests. Journal of Ecology 6:(in press).
Holdridge, L.R. 1967. Life Zone Ecology. Tropical Science Center, San Jose, Costa Rica.
Lugo, A.E. 1988. The future of the forest. Environment 30(7): 16-20, 41-45.
Lugo, A.E. Comparison of four small tropical tree plantations fPinus caribaea and Swietenia macrophvlla^ with
secondary forests of similar age. In review, Ecological Monographs. ,
Lugo, A.E., D, Wang, and F.H. Bormann. 1990. A comparative analysis of biomass production in five tropical
tree species. Forest Ecology and Management 31: (in press).
Lugo, A.E., E. Cuevas, and M J. Sanchez. 1990, Nutrient and mass in litter and top soil of ten tropical tree
plantations. Plant and Soil (in press).
Lugo, A.E. and F.N. Scatena. 1991. Ecosystem-level properties of the Luqillo Experimental Forest with
emphasis on the tabonuco forest. In A.E. Lugo and C. Lowe, eds., One Hundred Years of Tropical Forestry
Research. Springer-Verlag (in review).
Lugo, A.E., M J. Sanchez, and S. Brown. 1986. Land-use and organic carbon content of some subtropical soils.
Plant and Soil 96:185-196.
Wang, D., F.H. Bormann, A.E. Lugo, and R.D, Bowden. The implications of nutrient-use efficiency on fiber
harvest for five tropical tree taxa. Manuscript.
Weaver, P.L., RA. Birdsey, and A.E. Lugo. 1987. Soil organic matter in secondary forests of Puerto Rico.
Biotropiea 19:17-23.
77
-------
APPENDIX D: LETTER FROM DAVID JENKINSON
ROTHAMSTED
EXPERIMENTAL STATION
HARPENDEN ¦ HERTS ¦ AL5 2JO
Tel 0582 763133 • Fox: 0582 760981 ¦ Tele* 825726 REXP5T G
Acting Head a' Soils and Agronomy Department
ProfeuorJ.A. Coll, D.Sc,
Dr Mark G, Johnson
U.S. EPA Environment Research Laboratory
200 S.W, 35th Street
Corvallis
Oregon 97333
USA
8 February 1990
Dear Dr Johnson,
I am very sorry I will not be able to attend the workshop you are
arranging on 'Sequestering Carbon in Soils', particularly as I am sure it
will prove to be a roost interesting and useful meeting. When we, spoke on
the 'phone, I promised you details of Rothamsted work that might be
relevant to some of the workshop topics. Herewith a list.
1. Retention of organic manure carbon by soil
Two of the Rothamsted Classical Experiments provide data on this - see
enclosed Intecol paper, figures 5 and 6. A very large input of farmyard
manure (35 tonnes ha""' year-1 ) has been applied every year for more than
100 years in both, experiments. Thirty-five tonnes of wet FYM contains
about 3 tonnes of C, which comes from about 4 tonnes of fresh plant C,
or about 10 tonnes of fresh plant dry matter. In the Hoosfield experi-
ment (Fig. 6). 186 tonnes of FYM C had been added between 1851 and 1913.
In 1913 the FYM plot contained 49 tonnes more C than the unmanured
control, so that an average of 26% of the FYM C had been retained in the
soil over the 62 year period. The corresponding retention of FYM C in
the Broadbalk experiment (Fig. 5) over the 1843-1914 period was a little
less, 211. These figures are of course only averages - retention will
be a little higher in the early years and a little lower in the later
years of the period. As time goes on, the retention of C will decrease
as the soil comes to a new equilibrium, with the annual input of carbon
the soil, including that from FYH, balanced by the annual output of
COj-C,
2. Gain of C by soils reverting to woodland
I enclose two graphs showing the gains of organic C in the topsoil (0-23
cm) in two small areas of old arable land on Rothamsted Experimental
Farrr that were fenced off in the early 1880s and allowed to revert
naturally to woodland, I also enclose a reprint of an earlier paper
describing these two areas in detail. When sampled in 1964-65, some 80
years after reversion had started, the Ceescroft site had gained 21
tonnes C ha-1 in the 0-23 cm layer (corrected for changes in bulk
density) and 72 tonnes C in the trees: the corresponding figures for the
Broadbalk site were 43 tonnes and 110 tonnes.
[AWES AGRICULTURAL TRUST
78
-------
3• Crop yields and soil organic matter levels
1 enclose a reprint that may be of interest if you wish to consider this
particular aspect of the sequestration of organic C in soils.
I hope these comments will be of use and wish you well in your task of
organising the workshop.
With all good wishes
Yours sincerely
D,S, Jenkinson
Encs
79
-------
100
FYM
• c
u
4q;
u
"E
fz
?
O 20
j Unmanured
includes 10
tallow years
Predicted radiocmbon age 1280 years
~ Measured radiocBfbon age 1450 years
1880
Date
2000
1820
Fieure S Organic C in the top 23 cm of a Rothamsted soil under continuous winter wheat (Broad-
balk) Data calculated from in soil (as given by jenkinsor. 1977b, using a soil weight of 2.91
M kg''ha apart from the 1981 results which are from Powlsonet al. 1986), using C/N ratios of 9.4],
9 91 and 10 28 for plots 03, 08, and 22, respectively, and allowing for changes in bulk density in
pioi 22 Inorganic CEC is 10-8 me/100 g soil in this and in all other Rothamsted field experiments
(Figs. 6and7). Plot 03 is unmanured; the NPK plat <08) receives 144 kg N, 35 kg P, and 90 kg K/ha
per vear, apart from fallow years; the FY M plot (22) rece.ves 35t FYM annually. The FYM, ap-
olied in eariv autumn, was assumed to be equivalent to 75% of the original plant material from which
ft came and to contain DPM, RPM, and HUM (but no biomass) in the proportions 0.65,0.30, and
0 05 respectively In fallow'years, decomposition was assumed to proceed as usual with no fresh
FYM or plant debris entering the soil. The C inputs used, all in t C/ha per year, were: unmanured
plot, 1.2; NPK plot, 1.9; FYM plot, 1.9 (plant debris) + 3.0 (FYM).
100
80
60
U
o
'£
ra
Ds
o
40
20
FYM
FYM 1852-187 1,
nothing thereafter
Unmanured
5
Fallow
Fallow
r t
Fattow Fallow
1840
tsao
1920
Date
1960
2000
Fieure 6. Organic C in the top 23 cm of a soil under continuous spring barley (Hoosfield). Data,
corrected for changes in bulk density, are from Jenkinson and Johnston (1977) except for results
for 1982 (unpublished). Plot 1-0 is unmanured, the FYM plot (7-2) receives 351 FYM annually, the
FYM residues plot (7-i) received 351 FYM annually between 1852 and S871 and nothing since. The
FYM applied in late autumn (after 1916), was assumed to be equivalent to 60% of the plant material
from which it came and to contain DPM, RPM, and HUM (but no biomass) in the proportions 0.53,
0,38., and 0,09, respectively. Before 1916, FYM was applied in spring and was assumed to be
equivalent to 45®?o of the original plant material, the corresponding proportions being 0.34,0^49,
andO 17 The C irputs used, all in t C/ha per year, were unmanured plot, 1.1; FYM plot, 11.5 (plant
debris) +' 3.0 (FYM); FYM residues plot, as FYM plot during 1852-1871, 1.5 during 1872-1876.
thereafter, 1.1,
80
-------
Broadbalk Wilderness, Rothamsted Experimental Station.
100
80
•h m
o -C
« "***
o. w
O D
H C
c
H O
+-»
U
o
60
40
20
Predicted by Model4.
/\ Measured.
Model Inputs.
4.0 tonnes Qtia/yT
3 J tonnes Qha/yr
3.0 tonnes C/ha/yt
1810 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050
Time in Years.
,5
u
3
£
Geescroft Wilderness, Rothamsted Experimental Station.
90
so
70
60
JO
40
B0
20
10
0
—— Predicted by Model-.
>j< Measured.
Model Inputs.
3.5 lonries C/ha/yr
3.0 tonnes C/ha/yr
2 J tonnes C/hs/ys
2.0 tonnes Qha/yr
J i 1 ' I ' I
_L
i ¦ '
1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050
Time in Years.
81
-------
APPENDIX E; LETTER FROM HAMS JENNY
Memorandum on sequestering carbon in soils
Hans Jenny, February 1990
1. Soil Is a good preserver of organic carbon. In lower horizons
"apparent mean residence times" of 10,000 - 20,000 years have
been measured.
2. A world-wide estimate of soil carbon content (exclusive of
forest floor) was published by Jerry Olson and Paul Zinke. Dr.
W.M. Post may be familiar with it. The survey may tell where on
the globe soils acquired high carbon contents. I had concluded
that well-drained soils at high elevation in the tropics tend to
be high in carbon (H.J. The Soil Resource, p.320,19SC), and a
Safari to Mt. Kilimanjaro in Africa confirmed it.
3. To estimate past oxidation of soil carbon caused by farming,
assume the conservative loss of C for the soil depth of
0-20 cm, and 10^ for the rest of the profile.
k. Assumedly, incorporation of carbon into su'isoils occurs
mainly by root-growth and decay. Roots to depths of over 50 feet
have been observed.
5. With this in vien forests and grasslands and crop rotations
should be encouraged to include deep-rooting species.
6. This approach would require a reversal.cf the trend of plant
breeders who now direct the flow of photosynthate to stems, leaves ,
flowers and seeds. Scientists would have to start breeding root systems.
7. To fill up a soil with carbon to its carbon-carrying capacity
(near steady state) will require many generations of trees, shrubs ,
and other green species.
8. As a curiosity, if finely grounded basalt rock high in calcium
(CaO) were spread in forests it would fix C0£ and make a CaCOj -enriched
soil.
82
-------
APPENDIX F: WORKSHOP PARTICIPANTS
Dr. Chris Andersen
USEPA Environmental Research Lab.
200 SW 35th Street
Corvallis, OR 97333
Dr. Dominique Bachelet
NSI Technology Services
USEPA Environmental Research Lab.
200 SW 35th Street
Corvallis, OR 97333
Dr. Michael Bcare
Institute of Ecology
University of Georgia
Athens, GA 30602
Dr. Stan Buol
Department of Soil Science
North Carolina State University
Raleigh, NC 29650
Dr. Bruce Caldwell
Dept. of Microbiology
Oregon State University
Corvallis, OR 97331
Dr. Dale Cole
College of Forest Resources, AR-10
University of Washington
Seattle, WA 98195
Dr. Kerrait Cromack
Department of Forest Science
Forest Sciences Laboratory
Oregon State University
Corvallis, OR 97331 '
Dr. Richard Dick
Department of Soil Science
Oregon State University
Corvallis, OR 97331
Dr. John Duxbury
Dept. of Agronomy
Bradfield Hall
Cornell University
Ithaca, NY 14853
Dr. William Ferrell
Forest Science
Oregon State University
Corvallis, OR 97331
Dr. Cheryl Gay
NSI Technology Services
USEPA Environmental Research Lab
200 SW 35th Street
Corvallis, OR 97333
Dr. Robert Griffiths
Dept. of Microbiology
Oregon State University
Corvallis, OR 97331
Dr. Hermann Gucinski
NSI Technology Services
USEPA Environmental Research Lab
200 SW 35th Street
Corvallis, OR 97333
Ms. Kate Heaton
Bruce Company/USEPA
Office of Policy Analysis, PM-221
Climate Change Division
401 M Street, SW
Washington, DC 20460
Dr. Richard Houghton
The Woods Hole Research Center
P.O. Box 296
Woods Hole, MA 02543
Dr. El aine Ingham
Dept. of Botany & Plant Path.
Oregon State University
Corvallis, OR 97331
Dr. Dale Johnson
Biological Sciences Center
Desert Research Institute
P.O. Box 60220
Reno, Nevada 89506
Dr. Mark G. Johnson
USEPA Environmental Research Lab.
200 SW 35th Street
Corvallis, OR 97333
Mr. Jeffrey Kern
USEPA Environmental Research Lab.
200 SW 35th Street
Corvallis, OR 97333
Dr. Ellis Knox
USDA/Soil Conserv. Service
100 Centennial Mall N., Rm.345
Lincoln, NE 68508-3866
83
-------
Dr. Robert Lackey
USEPA Environmental Research Lab.
200 SW 35th Street
Corvallis, OR 97333
Dr. Rattan Lai
Department of Agronomy
The Ohio State University
2021 Coffey Road
Columbus, OH 43210-1086
Dr. Duane Lammers
USDA/Forest Science
USEPA Environmental Research Lab.
200 SW 35th Street
Corvallis, OR 97333
Dr. Leon Liegel
USDA Forest Service
USEPA Environmental Research Lab
200 SW 35th Street
Corvallis, OR 97333
Dr. Jeff Lee
USEPA Environmental Research Lab
200 SW 35th Street
Corvallis, OR 97333
Dr. Ariel Lugo
Institute of Tropical Forestry
USDA Forest Service
So. Forest Experiment Station
Call Box 2500
Rio Piedras, PR 00928-2500
Dr. Kim Mattson
USEPA Environmental Research Lab
200 SW 35th Street
Corvallis, OR 97333
Dr. David Myrold
Department of Soil Science
Oregon State University
Corvallis, OR 97331
Dr. William Parton
Natural Resource Ecology Laboratory
Colorado State University
Fort Collins, CO 80523
Dr. Dave Perry
Department of Forest Science
Forest Sciences Laboratory
Oregon State University
Corvallis, OR 97331
Dr. Charles Peterson
NSI Technology Services
USEPA Environmental Research Lab
200 SW 35th Street
Corvallis, OR 97333
Dr. W. Mac Posi
Environmental Sciences Division
P.O. Box 2008, Bldg. 1000
Oak Ridge National Laboratory
Oak Ridge, TN 37831-6335
Dr. Paul Rasmussen
USDA-ARS
Columbia Basin Agricultural Research Center
P.O. Box 370
Pendleton, OR 97801
Dr. Paul Rygiewicz
USEPA Environmental Research Lab
200 SW 35th Street
Corvallis, OR 97333
Dr. William Schlesinger
Department of Botany
Duke University
Durham, NC 27706
Mr. Paul Schroeder
NSI Technology Services
USEPA Environmental Research Lab
200 SW 35th Street
Corvallis, OR 97333
Mr. Paul Shaffer
NSI Technology Services
USEPA Environmental Research Lab.
200 SW 35th Street
Corvallis, OR 97333
Dr. Philip Sollins
Department of Forest Science
Forest Sciences Laboratory
Oregon State University
Corvallis, OR 97331
Dr. David Turner
NSI Technology Services
USEPA Environmental Research Lab.
200 SW 35th Street
Corvallis, OR 97333
84
-------
Dr. James Trappe
Department of Forest Science
Forest Sciences Laboratory
Oregon State University
Corvallis, OR 97331
Dr. Richard Waring
Department of Forest Science
Forest Sciences Laboratory
Oregon State University
Corvallis, OR 97331
Dr. Jack Winjum
Global Climate Team
USEPA Environmental Research Lab.
200 SW 35th Street
Corvallis, OR 97333
-------
_ TECHNICAL HEPORT DATA
(matt rmd Imwvtritmt on tht revtru befort eompttif*
216390
1. REPORT NO. 2.
EPA/600/3-91/031
I. R fD3 I-
4 TITLE AND SUBTITLE
Sequestering Carbon in Soil: A Workshop to
Explore the Potential for Mitigating Global
Climate Chancre
1. ACPjDRT. C ATE
April 1991
|U PERPORMINO ORGANIZATION CODE
J. AUTWORJSI
Mark G. Johnson and Jeffrey s. Kern
I. PSRPORMINO ORGANISATION REPORT NO.
1. PERPORMINB ORGANIZATION name AND ADDRESS
METI, Inc., ERL-Corvallis, OR
10. PROORAM ¦ IKMENT WO.
ft: e&wf aiunt1 "IB11:
13. SPONSORING AGENCY NAME AND ADDRESS
US Environmental Protection Agency
Environmental Research laboratory
200 SW 35th Street
Corvallis, OR 97333
1S TYPE Of REPORT AND PERIOD COVERED
14. »PONiiw«# jaftiiGf*e©tii
EPA/600/02
IB. supplementary notes
1991. U.S. Environmental Protection Agency, Environmental Research
Laboratory, Corvallis, OR.
IS. ABSTRACT
,
This workshop was an excellent forum for a scientific debate on the
potential of soils to sequester additional carbon from the atmosphere.
Two primary conclusions can be drawn from the workshop. First, that
steps should be taken to protect and preserve the size and integrity of
the global reservoir of soil carbon because continued losses of soil
carbon to the atmosphere could exacerbate global warming and climatic
change. Second, that steps should be taken to manage soils and
ecosystems to store additional carbon. The latter being accomplished
predominately by increasing net primary production. The major
uncertainties related to carbon sequestration in soils and specific
strategies for addressing these uncertainties and for managing soils to
store carbon were also identified at this workshop.
IT. KEY WORDS AND DOCUMENT ANALYSIS
1. DESCRIPTORS
t>.IDENT<*IEftS/OPEN ENDED TERMS
c. COSATi Fkld'Gjeup
carbon, climate change,
soils, global carbon cycle
IS. DISTRIBUTION STATEMENT
Release to Public
is,security ei,ass (TiuiMijxm/
UncTassmea
"t*?Ss
M SECURITY CLASS
Unclassified
22 PRICE
*PA p»rm U10-1 {S-J1)
------- |