soil carbon sequestration in tropical agroforestry systems: a feasibility appraisal

ENVSCI-699; No of Pages 13

Soil carbon sequestration in tropical agroforestry systems:a feasibility appraisal§

P.K. Ramachandran Nair a,*, Vimala D. Nair b, B. Mohan Kumar c, Solomon G. Haile a,b

aCenter for Subtropical Agroforestry, School of Forest Resources and Conservation, University of Florida, Gainesville, FL 32611-0410, USAb Soil and Water Science Department, University of Florida, Gainesville, FL 32611-0510, USAcDepartment of Silviculture and Agroforestry, College of Forestry, Kerala Agricultural University, Thrissur 680656, Kerala, India

e n v i r o n m e n t a l s c i e n c e & p o l i c y x x x ( 2 0 0 9 ) x x x – x x x

a r t i c l e i n f o

Keywords:

Clean development mechanism

Environmental services

Root biomass

Tree-based systems

Soil carbon storage

a b s t r a c t

Agroforestry is recognized as a strategy for soil carbon sequestration (SCS) under the

afforestation/reforestation activities, but our understanding of soil carbon (C) dynamics

under agroforestry systems (AFS) is not adequate. Although some SCS estimates are

available, many of them lack scientific rigor. Several interrelated and site-specific factors

ranging from agroecological conditions to system management practices influence the rate

and extent of SCS under AFS, so that generalizations tend to become unrealistic. Further-

more, widely and easily adoptable methodologies are not available for estimating the SCS

potential under different conditions. In spite of these, there is an increasing demand for

developing ‘‘best-bet estimates’’ based on the current level of knowledge and experience.

This document presents an attempt in that direction. The appraisal validates the conjecture

that AFS can contribute to SCS, and presents indicative ranges of SCS under different AFS in

the major agroecological regions of the tropics. The suggested values range from 5 to

10 kg C ha�1 in about 25 years in extensive tree-intercropping systems of arid and semiarid

lands to 100–250 kg C ha�1 in about 10 years in species-intensive multistrata shaded per-

ennial systems and homegardens of humid tropics.

# 2009 Elsevier Ltd. All rights reserved.

avai lab le at www.sc iencedi rec t .com

journal homepage: www.elsevier.com/locate/envsci

1. Introduction

Agroforestry has come of age forcefully during the past three

decades, when the production and sustainability attributes of

agroforestry systems (AFS) have received increasing recogni-

tion in a variety of circumstances (Nair, 2007; Nair et al., 2008).

Carbon (C) sequestration potential (CSP) of AFS is one of the

recent recognitions of the environmental benefits of this age-

old land-management approach. For example, the United

Nations Framework Convention on Climate Change (UNFCCC)

§ This paper is an abridged and modified version of a report preparedon Climate Change (UNFCCC). The UNFCCC, which holds intellectuaprovided consent to the use of the report as a basis for this paper. Thpersonal to the authors, do not reflect the views of UNFCCC and are

* Corresponding author. Tel.: +1 352 846 0880.E-mail address: [email protected] (P.K. Ramachandran Nair).

Please cite this article in press as: Nair, P.K.R., et al., Soil carbon seq

Environ. Sci. Policy (2009), doi:10.1016/j.envsci.2009.01.010

1462-9011/$ – see front matter # 2009 Elsevier Ltd. All rights reserveddoi:10.1016/j.envsci.2009.01.010

allows the use of C sequestration through afforestation and

reforestation (A & R) as greenhouse (GHG) offset activities

(http://unfccc.int/essential_background/glossary/items/

3666.php#C). Consequently, agroforestry (AF) became recog-

nized as a C sequestration activity under the A & R approach

(Albrecht and Kandji, 2003; Nair and Nair, 2003; Makundi and

Sathaye, 2004; Sharrow and Ismail, 2004; Haile et al., 2008;

Takimoto et al., 2008a; Nair et al., 2009). Since subsistence

farmers in developing countries are the major practitioners of

agroforestry, there is an added and attractive opportunity for

by the first author for the United Nations Framework Conventionl property rights and other proprietary rights to the report, hase information presented and/or views expressed in the paper arenot endorsed by UNFCCC.

uestration in tropical agroforestry systems: a feasibility appraisal.

.

http://unfccc.int/essential_background/glossary/items/3666.php

http://unfccc.int/essential_background/glossary/items/3666.php

mailto:[email protected]

http://dx.doi.org/10.1016/j.envsci.2009.01.010


e n v i r o n m e n t a l s c i e n c e & p o l i c y x x x ( 2 0 0 9 ) x x x – x x x2


them to benefit economically from agroforestry if the C

sequestered through AFS is sold to developed countries. This is

particularly relevant to the Clean Development Mechanism

(CDM) of UNFCCC, which allows industrialized countries with

a GHG reduction commitment to invest in mitigation projects

in developing countries as an alternative to what is generally

more costly in their own countries. Thus, the role of

agroforestry as a strategy for C sequestration has raised a

lot of expectations.

According to the Intergovernmental Panel on Climate

Change (IPCC, 2000), potential increases in carbon storage

may occur in agricultural and forest lands via (a) improved

management within a land use, (b) conversion of land use to

one with higher carbon stocks, or (c) increased carbon storage

in harvested products. In both (a) and (b) of the above, soils

play a vital role. Indeed, soils play a major role even in the

global C cycle: the soil C pool comprises soil organic C (SOC)

estimated at 1550 Pg (1 petagram = 1015 g = 1 billion tonnes)

and soil inorganic C about 750 Pg, both pools to 1-m depth

(Batjes, 1996). This total soil C pool of 2300 Pg is three times the

atmospheric pool of 770 Pg and 3.8 times the vegetation pool of

610 Pg; a reduction in soil C pool by 1 Pg is equivalent to an

atmospheric enrichment of CO2 by 0.47 ppmv (Lal, 2001). Thus,

any change in soil C pool would have a significant effect on the

global C budget.

Several authors have suggested that AFS have higher

potential to sequester C than pastures and field crops

(Sanchez, 2000; Roshetko et al., 2002; Sharrow and Ismail,

2004; Kirby and Potvin, 2007). This is based on the notion that

tree incorporation in croplands and pastures would result in

greater net sequestration of C both above- and below ground

(Palm et al., 2004; Haile et al., 2008; Nair et al., 2009). Validation

of this conjecture involves synthesizing available technical

and scientific data, which could lead to the outcomes of a

policy scenario. Several reports and estimates on C seques-

tration in AFS are available (for example, under the keywords

‘‘agroforestry’’ + ‘‘carbon sequestration’’, CAB Abstracts

(http://www.cabi.org) lists 266 papers, most of which were

published during the past 15 years). But, most of them are

based on the preconceived notion that AFS inherently have

high potential for soil C sequestration; process-oriented

investigations and rigorous scientific data are scanty, if not

non-existent. At the same time, the scientific community is

under increasing pressure from development agencies and

policy planners to develop ‘‘best-bet estimates’’ based on the

current level of knowledge and experience. This paper

presents such an appraisal; based on a critical assessment

of the available information on soil C sequestration rates in

AFS, it attempts to make some ‘‘bold’’ predictions on the SCS

potential in a range of AFS in the tropics.

2. Carbon sequestration potential ofagroforestry systems

2.1. Aboveground (vegetation) carbon sequestration

Since the focus of this paper is on soil C sequestration,

aboveground CSP is not discussed here in any detail. Never-

theless, because of the intricate and complex relationships



between aboveground and belowground C sequestration, it is

unrealistic to consider one in isolation from the other, which is

also the case with most reports on CSP of AFS. Briefly,

estimates of aboveground CSP are based on the assumption

that 45–50% of branch and 30% of foliage dry weight constitute

C (Shepherd and Montagnini, 2001; Schroth et al., 2002); these

estimates are highly variable, ranging from 0.29 to

15.21 Mg ha�1 yr�1. Since these estimates are directly related

to the estimated production potential of the system, they

depend on a number of factors including site characteristics,

land use types, species involved, stand age, and management

practices. Agroforestry systems in the arid, semiarid, and

degraded sites have a lower CSP than those in fertile humid

sites; and the temperate AFS have relatively lower vegetation

CSP than the tropical ones (Nair et al., 2009).

2.2. Belowground (soil) carbon sequestration

Carbon is sequestered in soils in two ways: direct and indirect

(Soil Science Society of America, SSSA, 2001). Direct soil C

sequestration occurs by inorganic chemical reactions that

convert CO2 into soil inorganic C compounds such as calcium

and magnesium carbonates. Indirect plant C sequestration

occurs as plants photosynthesize atmospheric CO2 into plant

biomass. Some of this plant biomass is indirectly sequestered

as SOC during decomposition processes. The amount of C

sequestered at a site reflects the long-term balance between C

uptake and release mechanisms. Because those flux rates are

large, changes such as shifts in land cover and/or land-use

practices that affect pools and fluxes of SOC have large

implications for the C cycle and the earth’s climate system. It

is estimated that historically soils account for about 58% of the

gross emission of CO2-C into the atmosphere from the

terrestrial ecosystems (Lal and Bruce, 1999; Lal, 2008).

The literature on soil carbon sequestration (SCS) potential

of AFS is scanty although rather plentiful reports are available

on the potential role of agricultural soils to sequester C.

Reviewing the SCS in AFS in comparison with other land-use

systems, Nair et al. (2009) noted a general trend of increasing

soil organic carbon (SOC), an indicator of SCS, in agroforestry,

and ranked the land-use systems in terms of their SOC content

in the order: forests > agroforests > tree plantations > arable

crops (Agroforests are complex multistrata systems, similar to

homegardens in structural complexity, but larger in size).

They further noted that the estimated values of SCS in AFS

that varied greatly were a reflection of the biophysical and

socioeconomic characteristics of the system parameters and/

or methodological artifacts.

The impact of any agroforestry system on soil C seques-

tration depends largely on the amount and quality of biomass

input provided by tree and non-tree components of the

system, and on properties of the soils, such as soil structure

and their aggregations. For example, in the establishment of

silvopastoral systems, some functional consequences are

inevitable when trees are allowed to grow in grass-dominated

land such as an open pasture. These include alterations in

above- and belowground total productivity, modifications to

rooting depth and distribution, and changes in the quantity

and quality of litter inputs (Connin et al., 1997; Jackson et al.,

2000; Jobbagy and Jackson, 2000). Such changes in vegetation


http://www.cabi.org/


Fig. 1 – A schematic representation of the holistic interrelationships among various major groups of factors that affect soil

carbon sequestration in agroforestry systems. The groups are so closely interconnected among each other that the

percentage contribution of each could be highly variable depending on the extent of influence by the other groups. Similar

to the Liebig’s law of minimum, the final output (soil carbon sequestration) will be conditioned by the factor that is most

limiting. The scheme is applicable to agroecosystem productivity in general; but the relative impact of management factors

could be less on soil C sequestration than on other aspects of ecosystem performance.

e n v i r o n m e n t a l s c i e n c e & p o l i c y x x x ( 2 0 0 9 ) x x x – x x x 3


component, litter, and soil characteristics modify the C

dynamics and storage in the ecosystem (Schlesinger et al.,

1990; Ojima et al., 1999).

Soil organic carbon contains a variety of fractions that

differ in decomposability and are very heterogeneous in

structure. The turnover of SOC is intimately linked with

organic matter quality (Agren et al., 1996; Martens, 2000).

Distinctive components of SOC have different residence times,

ranging from labile to stable forms (Carter, 1996). This concept

has led to the suggestion that SOC can be viewed as having an

active, labile pool (mean residence times [MRT’s] � 1–2 yr), a

slow pool (MRT’s � 25 yr), and a passive, recalcitrant pool

(MRT’s � 100–1000 yr) (Parton et al., 1987; Jenkinson, 1990;

Schimel et al., 1994; Torn et al., 2005). Further, protection of

SOC by silt and clay particles is well established (Sorensen,

1972; Ladd et al., 1985; Feller and Beare, 1997; Hassink, 1997;

von Lutzow et al., 2006, 2007). It is also known that aggregation

increases in less disturbed systems and that organic material

within the soil aggregates, especially the microaggregates,

have lower decomposition rate than those located outside the

aggregates (Elliott and Coleman, 1988; Six et al., 2000).

Thus, SCS in agroforestry systems – indeed in any land-use

system – is dependent on a large number of factors, ranging

from agroecological conditions to management practices. It is

important to keep in focus a holistic view of the interrelation-

ships among various factors for understanding the complex-

ities of soil C sequestration. With that objective, we suggest a

general schematic presentation showing the influence of

major factors, grouped into five categories, on soil C



sequestration (Fig. 1). These groups of factors are so closely

interconnected among each other that the percentage con-

tribution of each could be highly variable depending on the

extent of influence of each group of factors on one another.

Similar to the Liebig’s law of minimum, the final output (SOC)

will be conditioned by the factor that is most limiting.

Understandably, such a scheme could be applicable not only

to soil C sequestration, but to the overall agroecosystem

productivity; but the relative impact of management factors

could be less on soil C sequestration than on other aspects of

ecosystem performance.

3. Difficulties in estimating soil carbonsequestration in agroforestry systems

3.1. Area under agroforestry: present and potential

The first step in planning for capitalizing on the potential

benefits of any land-use system and designing development

plans for its sustainable utilization is to have an unambiguous

estimate of the area under the system at a given time period.

The lack of such an estimate is a major difficulty, not only in

estimating its CSP but in any activity aimed at the develop-

ment and utilization of the services and products of

agroforestry. Simply put, area statistics of agroforestry

systems are not available. Montagnini and Nair (2004) noted

that with no reliable estimates on the extent of area and the

gross variability expected in terms of tree species, stocking





levels and soil attributes, it is an ‘‘almost insurmountable’’

task to estimate C stocks in agroforestry.

Some attempts have been made to estimate the area and C

stocks under AF on a global scale. Dixon (1995) estimated the

area as 585 M to 1215 M ha, whereas Watson et al. (2000)

estimated that figure as 400 M ha. Based on the spread of AFS

under various geographical and ecological regions and

evaluation of previous reports, Nair et al. (2009) suggested

the area under agroforestry globally as more than 1000 M ha.

However, from the point of view of GHG mitigation, the extent

of area that could potentially be brought under agroforestry is

more relevant than the area currently under agroforestry,

because the potential of existing AFS to sequester additional C

in soil is perhaps limited, whereas new plantings offer better

scope in this regard (Takimoto et al., 2008b). Some estimates

are available on such areas that could be brought under

agroforestry. The IPCC (2000) estimated that 630 M ha of

unproductive croplands and grasslands could be converted to

agroforestry worldwide, with the potential to sequester

391 Gg C yr�1 by 2010 and 586 Gg C yr�1 by 2040. Another

estimate suggests that about 1900 M ha of land is degraded

due to erosion, salinity, fertility depletion, and advancing

deserts (Brown, 2004), and the potential of agroforestry to

reduce the hazards of erosion and desertification as well as to

rehabilitate such degraded land and to conserve soil and water

has been well recognized (Lal, 2004; Nair, 2007). Quoting IPCC

(2000) reports, Roshetko et al. (2007) suggest that globally the

greatest potential area for expanding agroforestry practices

and other forms of land-use intensification is in areas

considered ‘degraded’ at the margins of the humid tropics,

such as many secondary forest fallows, Imperata grasslands,

and degraded pastures. Verchot et al. (2007) estimated that in

the short-term (2008–2012), up to 5.3 M ha would be available

in developing countries for A & R under the CDM.

3.2. Root biomass studies in agroforestry systems

Roots are an important part of the soil C balance, because they

store large amounts of C in the soil. More than a third of the C

assimilated by the plant is eventually transported below

ground via root growth and turnover, root exudates (of organic

substances), and litter deposition. Root biomass in ecosystems

is often estimated from root-to-shoot ratios. The ratio ranges

from 0.18 to 0.30, with tropical forests in the lower range and

the temperate and boreal forests in the higher range (Cairns

et al., 1997). Depending on rooting depth, a considerable

amount of C is stored below the plow layer and is, therefore,

better protected from disturbance, which leads to longer

residence times in the soil. Available information on root

biomass in AFS (Table 1) indicates that root C inputs can

indeed be substantial in AFS. Woomer and Palm (1998)

reported that roots in tropical agroforestry systems could

have a time-averaged C stock ranging from about 6 Mg C ha�1

for shifting cultivation to about 20 Mg C ha�1 for tree fallows in

the top 0–50 cm soil depth.

As Schroth et al. (2007) have articulated, root dynamics are

one of the least understood aspects of system productivity and

functioning in AFS. Unlike in the case of annuals where all

roots die at the final harvest of the aboveground part, the

standing root system (the difference between cumulative



growth and cumulative decay at a given time) changes little in

the case of perennials, yet substantial growth and decay occur

simultaneously. This turnover through dead and sloughed-off

roots represents a substantial input to soil C pool; but the fine-

root dynamics are one of the least understood aspects of plant

life (Strand et al., 2008). Most of the biomass of the roots of

annual crops/grasses consists of fine roots (<2 mm in

diameter) whereas biomass of tree roots, which is a large

proportion of the belowground productivity, consists of coarse

roots (>2 mm diameter) (Albrecht et al., 2004; Akinnifesi et al.,

2004). Fine roots of both trees and crops have a relatively fast

turnover (days to weeks) (van Noordwijk et al., 1998), but the

lignified coarse roots decompose much more slowly and may

thus contribute substantially to belowground C stocks

(Vanlauwe et al., 1996).

Typical methods for root studies include spatially dis-

tributed soil cores or pits for fine and medium roots and partial

to complete excavation and/or allometry for coarse roots.

Often the distinction is not made between live and dead roots

and sampling depths are not standardized (Brown, 2002). The

amounts of fine roots, litterfall, pruning residues, and crops

are also variable, making it unreliable to use the general ratios

used for stem-biomass estimation. Although we have seen

some progress in the quantification of these issues at the plant

and ecosystem levels, most of the methods used to measure

such dynamics are problematic, especially in the case of

perennials (van Noordwijk et al., 2004).

Notwithstanding these difficulties and consequent lack of

rigor in most of the reported belowground biomass studies,

available reports compiled in Table 1 suggest that as much as

33% of the global annual net primary productivity (NPP) is used

for fine root production, which therefore is a major input to

soil organic matter pool (Nair et al., 1999). It is not clear,

however, if the 33% is an artifact of the method used in the

estimation of root biomass, given that in several cases a third

of the total productivity is considered to be belowground

productivity.

3.3. Lack of standard methods and procedures

In general, the methodologies and approaches used for

estimating C sequestration potential of land-use systems

have not been well developed (Nair et al., 2009). It is customary

that the reports and estimates on the subject start with

caveats and disclaimers that the data are inconsistent,

methods are not rigorous, and therefore the results and

estimates are inconclusive, and so on. Studies from which

these datasets are reported were mostly sporadic and short-

term, and not part of any unified or coordinated research

project. To make the situation even more difficult, the

background information reported for most of the studies is

inadequate and insufficient to even attempt to standardize the

datasets.

Since C sequestration refers to removing C from the

atmosphere and depositing it in a reservoir CSP should ideally

be reported as rates (mass per units of area and time). The

available data, however, are reported mostly as stocks (Nair

et al., 2009). This is understandable, given that time-sequence

studies on soil C involving intervals of several years are rare in

land-use systems in general and nonexistent in AFS. There-



Table 1 – Reported values of root (and aboveground) carbon (C) stocks in some tropical agroforestry systems.

Agroforestry/land-use system Major tree species Stand characteristicsa Method of estimation C stockb (Mg ha�1) Source

Root Above-ground

AF woodlot (Fodderbank,

fulewood lot); Kerala, India

Acacia auriculiformis Stand age: 9 yr; 2500

trees ha�1

Root excavation; only coarse

roots (>1.4 cm in diameter)

included

8.87 172 Kumar et al. (1998a)

Ailanthus triphysa 3.7 24.0

Artocarpus heterophyllus 5.07 46.1

Artocarpus hirsutus 5.58 35.0

Casuarina equisetifolia 2.80 50.6

Leucaena leucocephala 1.61 13.0

Paraserianthes falcataria 6.89 98.6

Phyllanthus emblica 6.32 40.7

Pterocarpus marsupium 3.65 36.7

Woodlot; Puerto Rico Casuarina equisetifolia Stand age: 4 yr; initial

spacing 1 m � 1 m;

planted in checkerboard

pattern in mixed stand

Total C stock excludes

litter and soil stocks

10.5 58.7 Parrotta (1999)

Eucalyptus robusta 6.9 36.2

Luecaena leucocephala 6.0 38.4

Mixed stands; Puerto Rico Eucalyptus + Casuarina 9.4 53.3

Casuarina + Leucaena 10.6 66.6

Eucalyptus + Leucaena 4.5 50.0

Cacao agroforests; Mekoe,

Cameroon

Cacao (Theobroma cacao)

in mixed stands

Stand age: 26 yr Sum of all roots; excludes

understory and litter stocks

20.5 145 Duguma et al. (2001)

Homegardens; Indonesia 45 tree species Stand age: 13 yr Excludes litter, herb,

and soil C stocks

8.8 44.1 Roshetko et al. (2002)

Silvopasture; Kurukhetra, India Acacia nilotica, Dalbergia sissoo,

Propspis juliflora; understory:

Desmostachya bipinnata;

Sporobolous marginatus

Sodic soils; stand age: 6 yr Excludes litter, herb,

and soil C stocks

Range: 2.03–4.81 Range: 7.72–14.63 Kaur et al. (2002)

Agrisilviculture; Chhattisgarh,

Central India

Gmelina arborea + eight

field crops

Stand age; 4 yr; trees were

fertilized and irrigated

during early 2 years

Coarse roots (>2 mm)

excavated; fine (<2 mm)

roots determined by soil cores

1.28 6.3 Swamy et al. (2003),

Swamy and Puri (2005)

Planted tree fallows; highlands

of eastern Kenya

Sesbania sesban Stand age: 2 yr; stand

density: 10 000 trees ha�1

Roots (<2 and >2 mm dia)

and root nodule mass

(up to 210 cm soil depth)

5.55 21.3 Stahl et al. (2002)

Calliandra calothyrsus 7.76 10.0

Grevillea robusta 8.85 25.1

Eucalyptus saligna 9.55 31.3

Agroforest; Ipetı-Embera, Panama Home and outfield gardens Excludes, litter, herb, and

soil C stocks

18.0 93.0 Kirby and Potvin (2007)

a Age and stand density details as reported in the original works are summarized; the sampling methods, however, were highly variable among the selected studies.b Wherever biomass values had been reported, the C stocks were deduced as 50% of the biomass stocks.

en

vir

on

me

nt

al

sc

ie

nc

e&

po

lic

yx

xx

(2

00

9)

xx

x–

xx

x5

EN

VS

CI-6

99

;N

oo

fP

ages

13

Please

citeth

isarticle

inp

ressas:

Nair,

P.K.R

.,et

al.,S

oil

carbo

nseq

uestratio

nin

trop

icalag

rofo

restrysy

stems:

afeasib

ilityap

praisal.

Env

iron

.S

ci.P

olicy

(20

09

),d

oi:1

0.1

01

6/j.en

vsci.2

00

9.0

1.0

10




fore, C stock estimates are taken as ‘‘surrogates’’ or indicators

for CSP. Vegetation C pools are sometimes linked to rotation

length of trees especially in the case of plantations and

therefore could be considered as rates; yet many of them are

estimated from and listed as C stocks.

A major challenge facing C stock estimation is accurate

estimation of tree biomass. One of the main technical issues is

the determination of the inventory and monitoring of stocks of

C sequestered in current and potential land uses and

management approaches; but a standard set of methods

and procedures is not available for this. Aboveground

biomass, based on which root biomass is often estimated, is

typically estimated using allometric equations developed for

trees in the natural forests. But, they generally lack accuracy

either because of their very location-specific- or much-

‘‘generalized’’ nature (Kumar et al., 1998a). The size of

individual tree canopies in an agroforestry setting also could

be different from that in a forest or single-species stand. In

addition, the crown and root architecture and tree manage-

ment practices are different; the resultant variations in

structure could result in erroneous estimates, especially if

generalized regression equations are used.

Direct methods and approaches used to quantify the

amount of C in forests are generally based on permanent

sample plots laid out in statistically sound designs (Mon-

tagnini and Nair, 2004). But indirect methods of estimation

such as remote sensing and modeling are subject to the

problem with simplified ‘‘average’’ or ‘‘default’’ values that are

often used as the bases for further computations. In GHG

mitigation projects, for example, it is not uncommon to find

differing ‘‘definitions’’ and interpretations of source and sink

categories and uncertainties about the basic processes leading

to emissions and/or removals. To estimate the effects of

harvest on C stocks, accurate information is required on three

items: pre-harvest biomass, the fraction of this biomass

harvested or damaged, and the fraction of the harvested

biomass removed; much of these, however, are not available

(de Jong, 2001).

Another point of uncertainty arises from the stand age.

Older stands generally store more C than younger ones; but

their ability to accumulate additional stock is minimal;

however, there is no consensus on this. Woomer et al.

(2000) argued that C accumulation rates (vegetation, soil, and

litter) for different land-use systems were not based so much

on the time a system re-grew, as on the land-use type. Fallows

established after initial cropping accumulated C stocks similar

to that of the original forest after about 20 years, but bush

fallows and multistrata agroforests accumulated about 60% of

initial forest C stocks in about 30 years, and pasture/grassland

after slash-and-burn resulted in continued gradual decline.

Conversely, a logistic model of biomass accumulated during

the first 15 years after abandonment in shifting cultivation

fallows showed that biomass production reached maximum

after 6 yr at 47 Mg dry matter ha�1 (Jepsen, 2006). Takimoto

et al. (2008b) suggested that, in the West African Sahel,

existing parkland systems (intercropping under scattered

trees such as Faidherbia albida and Vitelleria paradoxa: Boffa,

1999) had relatively low potential to sequester additional C,

whereas new tree-plantings as in fodder banks and live fences

in agricultural lands offered significant scope for sequestering



additional C. This is the logic used for promoting afforestation

projects as a strategy for C sequestration. One thing that

becomes clear is that extreme caution has to be exercised

when using the available information for developing default

values for soil C sequestration potential of AFS.

3.4. Other methodological issues

Considerable variations exist in the depth to which the soils

are sampled for soil C sequestration studies (in fact, soil

studies in general); this makes comparisons impossible. Soil C

sequestration studies, in general, are limited to the surface soil

(<20 cm); AFS are no exception: most SCS studies showing

improvements of SOC in AFS have concentrated on changes in

the topsoil layer, 0–20 cm, where the largest C pools are

detected (Makumba et al., 2007; Oelbermann and Voroney,

2007). Information is lacking on stocks of organic C in the

deeper soil layers where substantial amounts of tree roots

occur that could supply substantial amounts of C through root

exudates and root turnover. It is quite appropriate that calls

and recommendations are being made lately that future

studies in SOC sequestration be done by analyzing the soil

profile to about a one-meter depth rather than the surface

layer only (e.g., Blanco-Canqui and Lal, 2008).

4. Influence of tree species and tree-management practices on soil carbonsequestration

The tree species and the way they are combined in different

agroforestry systems will influence both the quantity and

quality of the biomass returned to the soil. Although

polycultures accrue more soil C, soil C stocks can also

increase under monocultures of trees, depending on the

species (Russell et al., 2004). The ‘‘native vs. exotic’’ species

controversy is among widely debated-but-not-yet-resolved

biological issues related to C sequestration by trees in AFS.

Much of such discussions originate from reports on C

sequestration in tree plantations (e.g., Nabuurs et al., 2000;

Zhang and Xu, 2003; Kaipainen et al., 2004), and in all these

reports, C sequestration is considered synonymous to C

stock and mostly of aboveground biomass—which in itself is

not fully correct as discussed earlier. Furthermore, it is still

unclear whether native species, because of their supposedly

better adaptability to local conditions, would be superior

to exotic ones for use in such plantations (Kumar et al.,

1998a).

Reports from plantations are nevertheless important to

AFS because many of the species promoted for plantations will

also be grown in AFS. Though not directly related to soil C

sequestration it is relevant to mention in this context that the

wood of slower-growing species is usually of higher specific

gravity than that of faster-growing species; therefore, the

slow-growing species may accumulate more C in the long-

term (Baker et al., 2004; Balvanera et al., 2005; Bunker et al.,

2005; Redondo-Brenes and Montagnini, 2006). The more

valuable, high-specific gravity wood also constitutes a longer

term sink for fixed C (e.g., construction timber, furniture, wood

crafts) than low specific-gravity wood used for short-lived





purposes such as packaging cases and poles. Therefore,

mixtures of fast-growing species and slower-growing species

that produce harvestable wood at different rotation times –

similar to species admixture in AFS – have been recommended

for Costa Rican plantations (Montagnini et al., 2005). The

impact of such mixed plantations on soil C sequestration per se

is not known. But, if the mixed plantations offer better

aboveground C sequestration, that in itself will be a positive

point in terms of their impact on overall GHG reduction.

Another group of mixed plantings is those involving N2-fixing

tropical species. Such plantings have been reported to produce

more aboveground biomass or volume production compared

to their monoculture stands (Kumar et al., 1998b; Bauhus et al.,

2004; Forrester et al., 2006); this could be expected to result in

corresponding increases in soil C sequestration as well. Resh

et al. (2002) found that soils beneath N2-fixing trees seques-

tered 0.11 kg m�2 y�1 more of total SOC compared with no

change under Eucalyptus. Many of the studies on encroach-

ment of woody plant species, mainly of N2-fixing tree legumes,

into natural grass systems showed significant increases in

SOC (Virginia and Jarrell, 1983; Bush and Van Auken, 1986;

Stock et al., 1995; Pugnaire et al., 1996). These increases are

attributed to greater inputs of N to N-limited ecosystems such

that N-limitation of plants and microbes is reduced leading to

greater plant productivity (Wardle, 1992). The impact of N2-

fixing tropical species on atmospheric concentration of GHGs

other than CO2 such as nitrous oxides (N2O) is frequently

mentioned (Firestone and Davidson, 1989; Sharkey and Loreto,

1993), but a solid body of research data is not yet available (Hall

and Asner, 2007).

Other silvicultural aspects such as stand density and

rotation length may also influence biomass production (and

the perceived soil CSP) of species. Overall, high-density stands

sequester larger amounts of C than lower-density stands.

Although these findings per se do not imply that mixed species

planting is not important, they suggest that choice of species

Table 2 – Summary of literature values on soil carbon stock u

Agroecological zones Major AF systems

Humid Lowlands Shaded perennial system

Alley cropping

Improved fallow

Homegardens

Tree intercropping

Silvopasture

Woodlots

Tropical Highlands Shaded perennial system

Silvopastoral systems

Arid and Semiarid Lowlands Silvopastoral systems

Fodder banks

Live fences

Grazing systems

Tree intercropping syste

Crop dominated

Fodder-tree dominated

Fuelwood dominated

a The soil depths and the study methods for different studies were highly

multiple literature sources. Specific literature citations are therefore not



and their management are critical to promoting C sequestra-

tion, both aboveground and belowground. This may, however,

create conflicts with plantation management objectives such

as timber, highlighting the need for stand density regulation

approaches that are in sync with land management objectives

(Kumar et al., 1995). Design of planting schemes to make trade-

offs between generating ecological services (e.g., C sequestra-

tion) and goods (e.g., timber) is indeed a major silvicultural

challenge.

5. Indicative estimates of soil carbonsequestration in agroforestry systems in thetropics

Notwithstanding the above-described deficiencies and gaps in

our knowledge and databases, we have attempted to make

some best-guess predictions on the extent of soil C seques-

tration in tropical agroforestry systems, primarily to meet the

demand for such estimates from development agencies, but

also to serve as an instigation to researchers for refinement of

the values. These estimates are based on about 150 peer-

reviewed articles and technical reports (only those cited in text

are included in the bibliography; for a more detailed literature

bibliography, see Nair et al., 2009), and supported by the

authors’ extensive research experiences in agroforestry and

soil carbon sequestration. Table 2 lists the reported range of

values of soil C stock in major agroforestry systems under

three major ecological regions in the tropics: humid lowlands

(<1500 m elevation), tropical highlands (>1500 m), and arid

and semiarid lowlands. No reported values, nor any close

approximations, could be found for some of the major

systems. Using Table 2 as a starting point, Table 3 was

prepared to present the projected values of soil C sequestra-

tion, taking into consideration all the basic elements identified

in the CDM recommendations.

nder tropical agroforestry systems.

Reported values of soil C stock (Mg ha�1)a

s 21–235

10–25

123–149

108–119

27–78

96–173

61–75

s 21–97

132–173

33

24

27–33

ms

20–70

25–80

30–90

variable. The listed range of values for each system is compiled from

given for each; literature citations can be found in Nair et al. (2009).



Table 3 – Indicative values of soil carbon stock and sequestration potential under major agroforestry systems in thetropicsa.

Major ecological regions andagroforestry systems

System characteristics—E:existing; N: new plantings;

TD: tree density (trees ha�1);age: years (yr)

Soil carbon (Mg C ha�1)b Time frame forrealizing the

potential (yr)cStock to 50 cm

soil depthPotential for

sequestering additionalC to 100 cm soil depth

Humid Lowlands

Shaded perennial systems E > 15 yr 100–200 20–30 10

N/young, <5-yr-old 70–150 100–200

Alley cropping E > 5 yr 20–45 25–75 >5

N or young < 5 yr 20–70 30–120 >10

Improved fallow 60–100 80–150

Homegardens Low TD < 750 trees/ha 60–90 70–150 >20

Medium TD > 750/ha 70–120 100–180 >20

Tree intercropping E, Low TD < 50/ha 20–80 50–100 >20

E, Med TD, 50–100/ha 40–100 70–120

E, High TD > 100/ha 50–120 80–150

Silvopasture (Grazing systems) E, TD Low, <25/ha 80–100 80–120 >20

E, TD High > 25/ha 80–120 90–150

Silvopasture (Fodder bank) E > 10-yr-old 60–95 30–60

N or young < 10 yr 75–95 50–150

Woodlots E > 10 yr 80–100 40–60 >20

N or young < 8 yr 50–80 50–150

Tropical Highlands

Shaded perennial systems E > 15-yr-old 100–200 20–50 10

N or young, <5 yr 70–150 100–250

Alley cropping E > 5 yr 30–60 40–70 >5

N or young < 5 yr 20–70 40–120 >10

Homegardens Low TD < 250 trees/ha 50–80 70–150 >20

Medium TD, >250/ha 70–150 100–200

Silvopasture (Grazing systems) E, TD Low, >20/ha 70–120 80–150 >20

E, TD High 80–150 90–160

Silvopasture (Fodder bank) E > 10 yr 60–100 30–70 >20

N or young < 8 yr 75–110 60–150

Woodlots E > 10-yr-old 80–100 40–70 >20

N or young < 5 yr 50–80 60–170

Arid and Semiarid Lands

(mostly lowlands)

Intercropping systems Parklands, W Afr Sahel E � 50 trees/ha 30–40 5–10 >25

Parklands, enrichment planting 20–30 30–50 >25

Silvopasture, semiarid regions:

Grazing systems

E � 50 trees/ha 30–40 5–10 >15

N: Planting trees in existing

grazing lands

20–30 30–50 >10

Fodder bank N 30–100

Fuelwood lot N

aThe values are ‘‘best-guess’’ estimates based on literature data (from nearly 150 peer-reviewed papers and reports) and authors’ experience.

Detailed literature citations are included in Nair et al. (2009). bThe soil C stock values are reported mostly from the upper soil layers, to less

than 50 cm depth. Therefore the estimates are for 0–50 cm soil depth. These as well as the values for sequestration potential will vary

enormously depending on a large number of site- and system-specific factors (see Fig. 1). b,cThe values proposed as potential for sequestering

additional C (column 4) are for up to 1 m depth considering the substantial amounts of tree roots and the SOC in deeper soil layers. It is

assumed that the existing systems have only limited potential in SCS unless they are significantly modified by management interventions

such as new (tree) planting and fertilization; but the potential could be substantial in new agroforestry initiatives. It is also recognized that

fairly long periods of time (column 5) are required to realize the potential for additional C sequestration in soils.



An apparent inconsistency in Table 2 is that the soil C

stock values are reported for 0–50 cm soil depth whereas the

C sequestration potential values are estimated for 0–100 cm

depth. This is because soil C stock values in literature are

reported mostly from the upper soil layers, to less than 50 cm

depth. But our recent research has shown the significant role

of tree roots in sequestering C in the deeper soil layers in AFS

under different agroecological conditions such as subtropi-

cal silvopasture in Florida, USA (Haile et al., 2008), parkland



and fodderbank systems in Mali (in the semiarid-to-arid

West African Sahel: Takimoto et al., 2008a,b,c), and home-

gardens in the humid lowlands of Kerala, India (Saha, 2008).

The suggested values of additional C sequestration potential

(Table 3, column 4) represent our best-bet estimates. They

are based on comparisons of datasets on soil C from different

soil layers in the above study sites and estimates of tree-root

contribution to SCS in deeper soil layers based on considera-

tions such as root-to-shoot ratio, root distribution in





different soil layers, and decomposition patterns of sloughed

off roots.

An underlying premise in this exercise is that the existing

systems have only limited potential in this regard unless they

are substantially modified by management interventions such

as new (tree) planting and fertilization; but the potential is

substantial in new agroforestry initiatives. It is also recognized

that fairly long periods of time (Table 3, column 5) are required

to realize the potential for additional C sequestration in soils.

This can be explained on the basis of the well-known growth

physiology of plants in general and trees in particular. During

the early stages of plant growth (which will be several years for

trees), most of the productivity goes into development of

storage organs such as roots and stem with very little of it

accreted back to soil towards build-up of soil C in organic

matter. Once the root systems have fully developed, more and

more of old roots will be sloughed off and newer ones

developed, the sloughed off roots forming a significant

addition to soil organic matter. Therefore, tree-mediated

sequestration of additional C to soil will be slow during the

early part of tree growth, but will pick up as the trees mature

until it levels off after its full maturity (following the

characteristic sigmoid growth curves of plants). This also

explains why existing (long-established) tree stands such as in

the parkland system of the West African Sahel have only

limited potential for sequestering additional soil C (Takimoto

et al., 2008b,c).

The reliability of the values presented in Table 3 for using

them as default values is a matter of judgment of the user. The

Table 4 – Major factors and conditions to be considered for seltropical agroforestry systems.

Factor/condition Effect

In genera

Management practices

Tillage and soil disturbance Inverse relation: less soil dist

C storage

Fertilizer/manure application More fertilizer application!

Plant residue management Leave more residue in soil!Nature of harvest Repeated harvest of woody p

C storage; repeated harvest o

(fodder, fuel, fruits) may not

System characteristics

Age of system Existing, long-established sys

limited potential for sequeste

Developmental stage of system Low SCS potential during ear

establishment; the length of t

Tree density More trees!more C storage

Species attributes Fast tree growth!more C st

production!more C storage

(i.e., more roots)!more C st

Ecological conditions

Rainfall (quantity and distribution) Higher and evenly distributed

Soil properties Higher clay + silt!more C st

sandy soil! less C storage; c

better soil conditions that

support tree growth!more C



projected values are only indicative. Since the values are

highly variable and subject to site-specific conditions, the user

has to exercise caution in selecting a value within the

suggested range for a given set of conditions. As a rule of

thumb, if the management practices are supportive of soil C

sequestration (for example: reduced tillage and soil distur-

bance, optimal fertilizer application for enhancing plant

growth, and judicious plant residue management ensuring

retention of as much plant residues on fields as possible), the

higher range of values could be used, and vice versa. Similarly,

adjustments must be made in relation to the development

stages of the AFS. Again, as a rule of thumb, the potential for

sequestering additional C in soil is rather limited in existing

‘‘mature’’ systems (such as the parkland systems of the West

African Sahel), whereas the potential is high when trees are

planted anew in treeless systems as part of afforestation

projects including new AF initiatives. The timeframe needed

to realize this potential in such new initiatives will again vary;

some indicative numbers (of years) are given in Table 3,

column 5. Some of these important factors and conditions that

need to be considered in selecting a ‘‘default’’ value for SCS of a

given AFS within the range suggested in Table 3 are

summarized in Table 4.

The values projected here will need to be validated against

field-research data. But, time sequence studies spanning over

several years are difficult to come by, at least in the foreseeable

future. Modeling is an approach that is being employed to

address similar issues of both biological and economic nature,

e.g., Ford-Robertson et al. (1999) and Robertson et al. (2004). But,

ecting default values for soil carbon sequestration (SCS) in

on soil carbon sequestration potential

l terms Relevance to specific systems

urbance!more Alley cropping and intercropping

systems

more C storage Production-oriented systems: alley

cropping, intercropping

more C storage Alley cropping, intercropping

erennials! less

f tree parts

impact soil C storage

Improved fallow, fodder banks,

multistrata systems

tems have

ring additional C

Parkland systems, multistrata

systems

ly stages of system

his early phase varies

Indicative values for different

systems are given in Table 3

Intercropping-, silvopastoral

(grazing)-, fodder bank-, and

multistrata systems

orage; high biomass

; higher shoot:root ratio

orage

All systems

rainfall!more C storage All systems

orage;

layey soil!more C storage;

storage





modeling is based on various assumptions and is subject to

vagariesoftheassumptionsmade.Nevertheless,modelinggives

usabroad-basedapproachtoaddressingtheissue.Noamountof

modeling or predictions, however, can be a substitute for

rigorousfieldresearch;atbest, theycan complement each other.

Then, there are total unknowns. For example, the extent of

variability in root-system proliferation of agroforestry trees

when grown in mixed (agroforestry) as opposed to monoculture

stands cannot be meaningfully guessed; it has to determined by

rigorous and meticulous research.

6. Concluding remarks

Agroforestry systems have a higher potential to sequester C

than pastures, or field crops, because tree incorporation in

croplands and pastures would result in greater net above-

ground as well as belowground C sequestration. Methodolo-

gical difficulties in estimating C stock of biomass and the

extent of soil C storage under varying conditions are serious

limitations in exploiting this low-cost environmental benefit

of agroforestry. Nevertheless, C trading is expanding globally,

and the CDM of the Kyoto Protocol offers an attractive

economic opportunity for subsistence farmers in developing

countries, the major practitioners of agroforestry, for selling

the C sequestered through agroforestry activities to indus-

trialized countries. It will be an environmental benefit to the

global community at large as well. The political environment

is also becoming favorable for enhancing smallholder invol-

vement in GHG mitigation projects.

r e f e r e n c e s

Agren, G.I., Bosatta, E., Balesdent, J., 1996. Isotopediscrimination during decomposition of organic matter: atheoretical analysis. Soil Sci. Soc. Am. J. 60, 1121–1126.

Akinnifesi, F.K., Rowe, E.C., Livesley, S.J., Kwesiga, F.R.,Vanlauwe, B., Alegre, J.C., 2004. Tree root architecture. In:van Noordwijk, M., Cadisch, G., Ong, C.K. (Eds.), Below-Ground Interactions in Tropical Agroecosystems: Conceptsand Models with Multiple Plant Components. CABIPublishing, Wallingford, UK, pp. 61–81.

Albrecht, A., Kandji, S.T., 2003. Carbon sequestration in tropicalagroforestry systems. Agric. Ecosyst. Environ. 99, 15–27.

Albrecht, G.C., Blanchart, E., Sitompul, S.M., Vanlauwe, B.,2004. Below-ground input: relationships with soil quality,soil C storage. In: van Noordwijk, M., Cadisch, G., Ong, C.K.(Eds.), Below-Ground Interactions in TropicalAgroecosystems: Concepts and Models with Multiple PlantComponents. CABI Publishing, Wallingford, UK, pp. 171–192.

Baker, T.R., Phillips, O.L., Malhi, Y., Almeida, S., Arroyo, L., DiFiore, A., Erwin, T., Killeen, T.J., Laurance, S.G., Laurance,W.F., Lewis, S.L., Lloyd, J., Monteagudo, A., Neill, D.A., Patino,S., Pitman, N.C.A., Silva, J.N.M., Vasquez Martinez, R., 2004.Variation in wood density determines spatial patterns inAmazonian forest biomass. Global Change Biol. 10, 1–18.

Balvanera, P., Kremen, C., Martinez-Ramos, M., 2005. Applyingcommunity structure analysis to ecosystem function:examples from pollination and carbon storage. Ecol. Appl.15, 360–375.



Batjes, N.H., 1996. The total carbon and nitrogen in soils of theworld. Eur. J. Soil Sci. 47, 151–163.

Bauhus, J., van Winden, A.P., Nicotra, A.B., 2004. Above-groundinteractions and productivity in mixed-species plantationsof Acacia mearnsii and Eucalyptus globulus. Can. J. For. Res. 34,686–694.

Blanco-Canqui, H., Lal, R., 2008. No-tillage and soil-profilecarbon sequestration: an on-farm assessment. Soil Sci. Soc.Am. J. 72, 693–701.

Boffa, J.M., 1999. Agroforestry parklands in sub-Saharan Africa.FAO Conservation Guide 34. FAO, Rome.

Brown, S., 2002. Measuring carbon in forests: current status andfuture challenges. Environ. Pollut. 116, 363–372.

Brown, L.R., 2004. Outgrowing the Earth: The Food SecurityChallenge in an Age of Falling Water Tables and RisingTemperatures. W.W. Norton, New York.

Bunker, D.E., DeClerk, F., Bradford, J.C., Colwell, R.K., Perfecto,Y., Phillips, O.L., Sankaran, M., Naeem, S., 2005. Species lossand above-ground carbon storage in a tropical forest.Science 310, 1029–1031.

Bush, J.K., Van Auken, O.W., 1986. Changes in nitrogen, carbon,and other surface soil properties during secondarysuccession. Soil Sci. Soc. Am. J. 50, 1597–1601.

Cairns, M.A., Brown, S., Helmer, E.H., Baumgardner, G.A., 1997.Root biomass allocation in the world’s upland forests.Oecologia (Berlin) 111, 1–11.

Carter, M.R., 1996. Analysis of soil organic matter inagroecosystems. In: Carter, M.R., Stewart, B.A. (Eds.),Structure and Organic Matter Storage in Agricultural Soil.CRC Press, Boca Raton, FL, pp. 3–11.

Connin, S.L., Virginia, R.A., Chamberlain, C.P., 1997. Carbonisotopes reveal soil organic matter dynamics following aridland shrub expansion. Oecologia 110, 374–386.

de Jong, B.H.J., 2001. Uncertainties in estimating the potentialfor carbon mitigation of forest management. For. Ecol.Manage. 154, 85–104.

Dixon, R.K., 1995. Agroforestry system: sources or sinks ofgreenhouse gases? Agroforest. Syst. 31, 99–116.

Duguma, B., Gockowski, J., Bakala, J., 2001. Smallholder cacao(Theobroma cacao Linn.) cultivation in agroforestry systemsof West and Central Africa: challenges and opportunities.

Elliott, E.T., Coleman, D.C., 1988. Let the soil work for us. Ecol.Bull. 39, 23–32.

Feller, C., Beare, M.H., 1997. Physical control of soil organicmatter dynamics in the tropics. Geoderma 79, 69–116.

Firestone, M.K., Davidson, E.A., 1989. Microbiological basis of NOand N2O production and consumption in soil. In: Andreae,M.O., Schimel, D.S. (Eds.), Exchange of Trace Gases BetweenTerrestrial Ecosystems and the Atmosphere. John Wiley &Sons Ltd., New York.

Ford-Robertson, J., Robertson, K., Maclaren, P., 1999. Modellingthe effect of land-use practices on greenhouse gasemissions and sinks in New Zealand. Environ. Sci. Policy 2,135–144.

Forrester, D.I., Bauhus, J., Cowie, A.L., 2006. Carbon allocation ina mixed-species plantation of Eucalyptus globulus and Acaciamearnsii. For. Ecol. Manage. 233, 275–284.

Haile, S.G., Nair, P.K.R., Nair, V.D., 2008. Carbon storage ofdifferent soil-size fractions in Florida silvopastoral systems.J. Environ. Qual. 37, 1789–1797.

Hall, S., Asner, G.P., 2007. Biological invasion alters regional N-oxide emissions from tropical rainforests. Global ChangeBiol. 13, 2143–2160.

Hassink, J., 1997. The capacity of soils to physically protectorganic C and N. Plant Soil 191, 77–87.

IPCC, 2000. Land Use, Land-use Change, and Forestry. A SpecialReport of the IPCC. Cambridge University Press, Cambridge,UK.





Jackson, N.A., Wallace, J.S., Ong, C.K., 2000. Tree pruning as ameans of controlling water use in an agroforestry system inKenya. For. Ecol. Manage. 126, 133–148.

Jenkinson, D.S., 1990. The turnover of organic carbon andnitrogen in soil. Philos. Trans. R. Soc. Lond. 329B, 361–368.

Jepsen, M.R., 2006. Above-ground carbon stocks in tropicalfallows, Sarawak, Malaysia. For. Ecol. Manage. 225, 287–295.

Jobbagy, E.G., Jackson, R.B., 2000. The vertical distribution of soilorganic carbon and its relation to climate and vegetation.Ecol. Appl. 10, 423–436.

Kaipainen, T., Liski, J., Pussinen, A., Karjalainen, T., 2004.Managing carbon sinks by changing rotation length inEuropean forests. Environ. Sci. Policy 7, 205–219.

Kaur, B., Gupta, S.R., Singh, G., 2002. Carbon storage andnitrogen cycling in silvopastoral systems on a sodic soil innorthwestern India. Agroforest. Syst. 54, 21–29.

Kirby, K.R., Potvin, C., 2007. Variation in carbon storageamong tree species: implications for the management ofa small-scale carbon sink project. For. Ecol. Manage. 246,208–221.

Kumar, B.M., Long, J.N., Kumar, P., 1995. A density managementdiagram for teak plantations of Kerala in peninsular India.For. Ecol. Manage. 74, 125–132.

Kumar, B.M., George, S.J., Jamaludheen, V., Suresh, T.K., 1998a.Comparison of biomass production, tree allometry andnutrient use efficiency of multipurpose trees grown in woodlot and silvopastoral experiments in Kerala, India. For. Ecol.Manage. 112, 145–163.

Kumar, B.M., Kumar, S.S., Fisher, R.F., 1998b. Intercropping teakwith Leucaena increases tree growth and modifies soilcharacteristics. Agroforest. Syst. 42, 81–89.

Ladd, J.N., Amato, M., Oades, J.M., 1985. Decomposition of plantmaterial in Australian soils. III. Residual organic andmicrobial biomass carbon and nitrogen from isotope-labeled legume material and soil organic matter,decomposing under field conditions. Aust. J. Soil Res. 23,603–611.

Lal, R., 2001. Potential of desertification control to sequestercarbon and mitigate the greenhouse effect. Clim. Change51, 35–72.

Lal, R., 2004. Soil carbon sequestration impacts on globalclimate change and food security. Science 304, 1623–1627.

Lal, R., 2008. Soil carbon stocks under present and futureclimate with specific reference to European ecoregions.Nutr. Cycl. Agroecosyst. 81, 113–127.

Lal, R., Bruce, J.P., 1999. The potential of world cropland soils tosequester C and mitigate the greenhouse effect. Environ.Sci. Policy 2, 177–185.

Makumba, W., Akinnifesi, F.K., Janssen, B., Oenema, O., 2007.Long-term impact of a gliricidia–maize intercroppingsystem on carbon sequestration in southern Malawi. Agric.Ecosyst. Environ. 118, 237–243.

Makundi, W.R., Sathaye, J.A., 2004. GHG mitigation potentialand cost in tropical forestry—relative role for agroforestry.Environ. Dev. Sustain. 6, 235–260.

Martens, D.A., 2000. Plant residue biochemistry regulates soilcarbon cycling and carbon sequestration. Soil Biol. Biochem.32, 361–369.

Montagnini, F., Nair, P.K.R., 2004. Carbon sequestration: anunderexploited environmental benefit of agroforestrysystems. Agroforest. Syst. 61, 281–295.

Montagnini, F., Cusack, D., Petit, B., Kanninen, M., 2005.Environmental services of native tree plantations andagroforestry systems in Central America. J. Sust. For. 21, 51–67.

Nabuurs, G.J., Dolman, A.J., Verkaik, E., Kuikman, P.J., vanDiepen, C.A., Whitmore, A.P., Daamen, W.P., Oenema, O.,Kabat, P., Mohren, G.M.J., 2000. Article 3.3 and 3.4 of theKyoto Protocol: consequences for industrialised countries’



commitment, the monitoring needs, and possible sideeffects. Environ. Sci. Policy 3, 123–134.

Nair, P.K.R., 2007. The coming of age of agroforestry. J. Sci. FoodAgric. 87, 1613–1619.

Nair, P.K.R., Nair, V.D., 2003. Carbon storage in North Americanagroforestry systems. In: Kimble, J., Heath, L.S., Birdsey,R.A., Lal, R. (Eds.), The Potential of U.S. Forest Soils toSequester Carbon and Mitigate the Greenhouse Effect. CRCPress, Boca Raton, FL, USA, pp. 333–346.

Nair, P.K.R., Buresh, R.J., Mugendi, D.N., Latt, C.R., 1999.Nutrient cycling in tropical agroforestry systems: mythsand science. In: Buck, L.E., Lassoie, J.P., Fernandes, E.C.M.(Eds.), Agroforestry in Sustainable Agricultural Systems.CRC Press, Boca Raton, FL, pp. 1–31.

Nair, P.K.R., Gordon, A.M., Mosquera-Losada, M.-R., 2008.Agroforestry. In: Jorgensen, S.E., Fath, B.D. (Eds.), EcologicalEngineering, of Encyclopedia of Ecology, vol. 1. Elsevier,Oxford, U.K., pp. 101–110.

Nair, P.K.R., Kumar, B.M., Nair, V.D., 2009. Agroforestry as astrategy for carbon sequestration. J. Plant Nutri. Soil Sci.172, 10–23.

Oelbermann, M., Voroney, R.P., 2007. Carbon and nitrogen in atemperate agroforestry system: using stable isotopes as atool to understand soil dynamics. Ecol. Eng. 29, 342–349.

Ojima, D.S., Kittel, T.G.F., Rosswall, T., Walker, B.H., 1999.Critical issues for understanding global change effects onterrestrial ecosystems. Ecol. Appl. 1, 316–325.

Palm, C.A., Tomich, T., van Noordwijk, M., Vosti, S., Alegre, J.,Gockowski, J., Verchot, L., 2004. Mitigating GHG emissions inthe humid tropics: case studies from the Alternatives toSlash-and-Burn Program, ASB. Environ. Dev. Sust. 6, 145–162.

Parrotta, J.A., 1999. Productivity, nutrient cycling and successionin single- and mixed-species stands of Casuarinaequisetifolia, Eucalyptus robusta and Leucaena leucocephala inPuerto Rico. For. Ecol. Manage. 124, 45–77.

Parton, W.J., Schimel, D.S., Cole, C.V., Ojima, D.S., 1987. Analysisof factors controlling soil organic matter levels in GreatPlains grasslands. Soil Sci. Soc. Am. J. 51, 1173–1179.

Pugnaire, F.I., Haase, P., Puigdefabregas, J., Cueto, M., Clark, S.C.,Incoll, L.D., 1996. Facilitation and succession under thecanopy of a legumnious shrub, Retama sphaerocarpa, in asemi-arid environment in south-east Spain. Oikos 76, 455–464.

Redondo-Brenes, A., Montagnini, F., 2006. Growth, productivity,aboveground biomass, and carbon sequestration of pureand mixed native tree plantations in the Caribbeanlowlands of Costa Rica. For. Ecol. Manage. 232, 168–178.

Resh, S.C., Binkley, D., Parrotta, J.A., 2002. Greater soil carbonsequestration under nitrogen-fixing trees compared withEucalyptus species. Ecosystems 5, 217–231.

Robertson, K., Loza-Balbuena, I., Ford-Robertson, J., 2004.Monitoring and economic factors affecting the economicviability of afforestation for carbon sequestration projects.Environ. Sci. Policy 7, 465–475.

Roshetko, M., Delaney, M., Hairiah, K., Purnomosidhi, P., 2002.Carbon stocks in Indonesian homegarden systems: cansmallholder systems be targeted for increased carbonstorage? Am. J. Alt. Agric. 17, 125–137.

Roshetko, J.M., Lasco, R.D., Delos Angeles, M.D., 2007.Smallholder agroforestry systems for carbon storage.Mitigation Adapt. Strategies Global Change 12, 219–242.

Russell, A.E., Cambardella, C.A., Ewel, J.J., Parkin, T.B., 2004.Species, rotation, and life-form diversity effects on soilcarbon in experimental tropical ecosystems. Ecol. Appl. 14,47–60.

Saha, S.K., 2008. Carbon sequestration potential of tropicalhomegardens and related land-use systems in Kerala, India.Ph.D. Dissertation. University of Florida, Gainesville, FL.





Sanchez, P.A., 2000. Linking climate change research with foodsecurity and poverty reduction in the tropics. Agric. Ecosyst.Environ. 82, 371–383.

Schimel, D.S., Braswell, B.H., Holland, E.A., McKeown, R., Ojima,D.S., Painter, T.H., Parton, W.J., Townsend, A.R., 1994.Climatic, edaphic, and biotic controls over storage andturnover of carbon in soils. Global Biogeochem. Cycles 8,279–293.

Schlesinger, W.H., Reynolds, J.E., Cunningham, G.L., Huenneke,L.F., Jarrell, W.M., Virginia, R.A., Whitford, W.G., 1990.Biological feedbacks in global desertification. Science 247,1043–1048.

Schroth, G., D’Angelo, S.A., Teixeira, W.G., Haag, D., Lieberei, R.,2002. Conversion of secondary forest into agroforestry andmonoculture plantations in Amazonia: consequences forbiomass, litter and soil carbon stocks after 7 years. For. Ecol.Manage. 163, 131–150.

Schroth, G., Schaller, M., Jimenez, F., 2007. Below-groundinteractions in tree-crop agroforestry: need for a newapproach. In: Batish, D.R., Kohli, R.K., Jose, S., Singh, H.P.(Eds.), Ecological Basis of Agroforestry. CRC Press, BocaRaton, FL, pp. 159–170.

Sharkey, T.D., Loreto, F., 1993. Water-stress, temperature, andlight effects on the capacity for isoprene emission andphotosynthesis of kudzu leaves. Oecologia 95, 328–333.

Sharrow, S.H., Ismail, S., 2004. Carbon and nitrogen storage inagroforests, tree plantations, and pastures in westernOregon, USA. Agroforest. Syst. 60, 123–130.

Shepherd, D., Montagnini, F., 2001. Carbon sequestrationpotential in mixed and pure tree plantations in the humidtropics. J. Trop. For. Sci. 13, 450–459.

Six, J., Elliott, E.T., Paustian, K., 2000. Soil macroaggregateturnover and microaggregate formation: a mechanism for Csequestration under no-tillage agriculture. Soil Biol.Biochem. 32, 2099–2103.

Soil Science Society of America, 2001. Carbon Sequestration:Position of the Soil Science Society of America (SSSA). SSSA,Madison, WI. Available at www.soils.org/pdf/pos_paper_carb_seq.pdf. Accessed on 10 November 2008.

Sorensen, L.H., 1972. Stabilization of newly formed amino acidmetabolites in soil by clay minerals. Soil Sci. Soc. Am. J. 114,5–11.

Stahl, L., Nyberg, G., Hogberg, P., Buresh, R.J., 2002. Effects ofplanted tree fallows on soil nitrogen dynamics, above-ground and root biomass, N2-fixation and subsequentmaize crop productivity in Kenya. Plant Soil 243, 103–117.

Stock, W.D., Wienand, K.T., Baker, A.C., 1995. Impacts ofinvading N2-fixing Acacia species on patterns of nutrientcycling in two Cape ecosystems: evidence from soilincubation studies and 15N natural abundance values.Oecologia 101, 375–382.

Strand, A.E., Pritchard, S.G., McCormack, M.L., Davis, M.A., Oren,R., 2008. Irreconcilable differences: fine-root life spans andsoil carbon persistence. Science 319, 456–458.

Swamy, S.L., Puri, S., 2005. Biomass production and C-sequestration of Gmelina arborea in plantation andagroforestry system in India. Agroforest. Syst. 64, 181–195.

Swamy, S.L., Mishra, A., Puri, S., 2003. Biomass production androot distribution of Gmelina arborea under an agrisilviculturesystem in subhumid tropics of Central India. New Forests26, 167–186.

Takimoto, A., Nair, P.K.R., Alavalapati, J.R.R., 2008a.Socioeconomic potential of carbon sequestration throughagroforestry in the West African Sahel. Mitigation Adapt.Strategies Global Change 13, 745–761.

Takimoto, A., Nair, P.K.R., Nair, V.D., 2008b. Carbon stock andsequestration potential of traditional and improvedagroforestry systems in the West African Sahel. Agric.Ecosyst. Environ. 125, 159–166.



Takimoto, A., Nair, V.D., Nair, P.K.R., 2008c. Contribution oftrees to soil carbon sequestration under agroforestrysystems in the West African Sahel. Agroforest. Syst.,doi:10.1007/s10457-008-9179-5.

Torn, M.S., Vitousek, P.M., Trumbore, S.E., 2005. The influence ofnutrient availability on soil organic matter turnoverestimated by incubations and radiocarbon modeling.Ecosystems 8, 352–372.

van Noordwijk, M., Hairiah, K., Woomer, P.L., Murdiyarso, D.,1998. Criteria and indicators of forest soils used for slash-and-burn agriculture and alternative land uses inIndonesia. In: Davidson, E. (Ed.), The Contribution of SoilScience to the Development and Implementation of Criteriaand Indicators of Sustainable Forest Management. ASASpecial Publication, Am Soc Agronomy, Madison, WI, pp.137–153.

van Noordwijk, M., Rahayu, S., Williams, S.E., Hairiah, K.,Khasanah, N., Schroth, G., 2004. Crop and tree root-systemdynamics. In: van Noordwijk, M., Cadisch, G., Ong, C.K.(Eds.), Below-Ground Interactions in TropicalAgroecosystems: Concepts and Models with Multiple PlantComponents. CABI Publishing, Wallingford, UK, pp. 83–107.

Vanlauwe, B., Nwoke, O.C., Sanginga, N., Merckx, R., 1996.Impact of the residue quality on the carbon and nitrogenmineralization of leaf and root residues of threeagroforestry species. Plant Soil 183, 221–231.

Verchot, L.V., Van Noordwijk, M., Kandji, S., Tomich, T., Ong, C.,Albrecht, A., Mackensen, J., Bantilan, C., Anupama, K.V.,Palm, V., 2007. Climate change: linking adaptation andmitigation through agroforestry. Mitigation Adapt.Strategies Global Change 12, 901–918.

Virginia, R.A., Jarrell, W.M., 1983. Soil properties in a mesquite-dominated Sonoran desert ecosystem. Soil Sci. Soc. Am. J.47, 138–144.

von Lutzow, M., Koegel-Knabner, I., Ekschmitt, K., Matzner, E.,Guggenberger, G., Marschner, B., Flessa, H., 2006.Stabilization of organic matter in temperate soils:mechanisms and their relevance under different soilconditions—a review. Eur. J. Soil Sci. 57, 426–445.

von Lutzow, M., Koegel-Knabner, I., Ekschmitt, K., Flessa, H.,Guggenberger, G., Matzner, E., Marschner, B., 2007. SOMfractionation methods: relevance to functional pools and tostabilization mechanisms. Soil Biol. Biochem. 39, 2183–2207.

Wardle, D.A., 1992. A comparative assessment of factors whichinfluence microbial biomass carbon and nitrogen levels insoil. Biol. Rev. 67, 321–358.

Watson, R.T., Noble, I.R., Bolin, B., Ravindranath, N.H., Verardo,D.J., Dokken, D.J., (Eds.), 2000. Intergovernmental Panel onClimate Change (IPCC). Land Use, Land-use Change, andForestry. A Special Report of the Intergovernmental Panelon Climate Change. Cambridge University Press,Cambridge, UK.

Woomer, P.L., Palm, C.A., 1998. An approach to estimatingsystem carbon stocks in tropical forests and associated landuses. Commonwealth For. Rev. 77, 181–190.

Woomer, P.L., Palm, C.A., Alegre, J., Castilla, C., Cordeiro, D.G.,Hairiah, K., Kotto-Same, J., Moukam, A., Ricse, A., Rodrigues,V., van Noordwijk, M., 2000. Slash-and-burn effects oncarbon stocks in the humid tropics. In: Lal, R., Kimble, J.M.,Stewart, B.A. (Eds.), Global Climate Change and TropicalEcosystems. Advances in Soil Science. CRC Press, BocaRaton, FL, pp. 99–115.

Zhang, X.-Q., Xu, D., 2003. Potential carbon sequestration inChina’s forests. Environ. Sci. Policy 6, 421–432.

P.K. Ramachandran Nair is a researcher and educator and a worldleader in agroforestry. He has been on the faculty at the Universityof Florida since 1987, where he is distinguished professor anddirector of the Center for Subtropical Agroforestry. Trained in


http://www.soils.org/pdf/pos_paper_carb_seq.pdf

http://www.soils.org/pdf/pos_paper_carb_seq.pdf

http://dx.doi.org/10.1007/s10457-008-9179-5

http://dx.doi.org/10.1007/s10457-008-9179-5




India (PhD), UK (postdoctoral) and Germany (second PhD), he hasreceived Doctor of Science honoris causa degrees from universitiesin Canada, Ghana, Japan, and Spain; recognitions by several pro-fessional societies worldwide; and numerous awards includingthe Humboldt Prize, Germany. He was one of the founders ofICRAF (World Agroforestry Centre), Nairobi, Kenya, where heworked for nearly 10 years. His current research areas includecarbon sequestration under agroforestry systems.

Vimala Nair is research associate professor, Soil and WaterScience Department, University of Florida, specializing in envir-onmental quality research with emphasis on nutrient/waste man-agement and carbon sequestration. She has received training inIndia (BSc and MSc), Germany (PhD) and the USA (post-doctoralassociate).



B. Mohan Kumar is associate dean and head of the College ofForestry, Kerala Agricultural University, Thrissur, India, where hehas been on its faculty since 1978. His current research interestsinclude sustainable land-use systems with special reference toagroforestry and the effects of forest management practices onecosystem processes, particularly nutrient cycling, soil fertility,and vegetation dynamics.

Solomon G. Haile is a postdoctoral associate at University of FloridaSoil and Water Science Department. His research interest is inbiogeochemical interface between plant communities, with empha-sis on environmental services of trees. His PhD study was on soilcarbon sequestration in tree-based pasture agroforestry systems inFlorida.Prior tothat,hewasonthe faculty of theDepartmentof PlantScience, University of Asmara, Eritrea for five years.



soil carbon sequestration in tropical agroforestry systems: a feasibility appraisal

Documents