Agriculture, Ecosystems and Environment 104 (2004) 359–377

Review

Carbon sequestration in tropical and temperate agroforestry systems: a review with examples from Costa Rica and southern Canada Maren Oelbermann a,∗ , R. Paul Voroney b,1 , A.M. Gordon c,2 a Department of Earth Sciences, University of Waterloo, Waterloo, Ont., Canada N2L 3G1 Department of Land Resource Science, University of Guelph, Guelph, Ont., Canada N1G 2W1 c Department of Environmental Biology, University of Guelph, Guelph, Ont., Canada N1G 2W1

b

Received 9 April 2003; received in revised form 2 April 2004; accepted 5 April 2004

Abstract Deforestation in the tropics, and fossil fuel burning in temperate regions contribute to the largest flux of CO2 to the atmosphere. Therefore, land-use systems that increase the soil organic matter (SOM) pool and stabilize soil organic carbon (SOC) need to be implemented. Agroforestry systems have the potential to sequester atmospheric carbon (C) in trees and soil while maintaining sustainable productivity. The potential to sequester C in agroforestry systems in tropical and temperate regions is promising, but little information is available to date. The objective of this paper is to give an overview of the history of agroforestry and to outline differences in management practices between tropical and temperate systems. This review focuses on C inputs, SOC pools and SOC stabilization with highlights from Costa Rican and Canadian systems, and their role in C sequestration and trading. The potential to sequester C in aboveground components in agroforestry systems is estimated to be 2.1 × 109 Mg C year−1 in tropical and 1.9 × 109 Mg C year−1 in temperate biomes. However, the type of agroforestry systems and their capacity to sequester C vary globally. For example, alley cropping is an agroforestry practice where trees are integrated with crops, therefore storing C in the woody components of the trees and in the soil, with a continual addition of organic material from tree prunings and crop residues. Studies from Costa Rica have shown that a 10-year-old system with E. poeppigiana sequestered C at a rate of 0.4 Mg C ha−1 year−1 in coarse roots and 0.3 Mg C ha−1 year−1 in tree trunks. Tree branches and leaves are added to the soil as mulch, contributing 1.4 Mg C ha−1 year−1 in addition to 3.0 Mg ha−1 year−1 from crop residues. This resulted in an annual increase of the SOC pool by 0.6 Mg ha−1 year−1 . Despite the two crop rotations in tropical agroforests, C input from crop residues is similar between the two biomes. The total organic matter input, however, is still greater in tropical systems due to the larger addition from tree prunings. This greater input does not necessarily increase the SOC pool significantly when compared to a temperate system of similar age as a result of faster turnover rates of the SOM pool. © 2004 Elsevier B.V. All rights reserved. Keywords: Alley cropping; Canada; Carbon sequestration; Carbon input; Costa rica; Soil carbon stabilization; Soil organic matter

∗

Corresponding author. Tel.: +1-519-888-4567x6495; fax: +1-519-746-7484. E-mail addresses: [email protected] (M. Oelbermann), [email protected] (R. Paul Voroney), [email protected] (A.M. Gordon). 1 Tel.: +1-519-824-4120x53057; fax: +1-519-824-5730. 2 Tel.: +1-519-824-4120x52415; fax: +1-519-837-0442. 0167-8809/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.agee.2004.04.001

360

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Historical perspectives on agroforestry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tropical and temperate agroforestry practices: examples from Costa Rica and Canada . . . . . . . Carbon dynamics in tropical and temperate agroforestry systems . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Carbon input from aboveground tree components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. The role of tree roots in agroforestry systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Input from crop residues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Decomposition and stabilization of organic matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5. Soil organic carbon pools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Linking agroforestry to carbon sequestration and carbon trading . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Conclusions and future research needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. 2. 3. 4.

1. Introduction Natural exchanges of carbon (C) between the atmosphere, the oceans and terrestrial ecosystems are currently modified by human activities. These changes are the result of fossil fuel burning in the northern hemisphere and the conversion of forests to agricultural land in the tropics (Paustian et al., 2000). Such activities have increased atmospheric C concentrations by 28% over the past 150 years (Schlesinger, 1997), resulting in an annual accumulation rate of 3.5 Pg C year−1 (Paustian et al., 2000). Some studies have suggested that elevated atmospheric CO2 concentrations could increase forest productivity (Saxe et al., 1998; Wang et al., 1998; Norby et al., 1999) and accelerate growth rates of agricultural crops and grasslands, leading to higher crop productivity and grain yield (Tubiello et al., 1999). However negative changes are also associated with a higher atmospheric CO2 concentration including an increase in global temperatures (Schlesinger, 1997; Costa and Foley, 2000). The concern over rising levels of CO2 and other greenhouse gases (GHGs) in the atmosphere was addressed at the third meeting of the United Nations Framework Convention on Climate Change (UNFCCC) in 1997, in Kyoto, Japan. The results of this convention led to an agreement, known as the Kyoto protocol, among participating countries to reduce the rising levels of CO2 , and other GHGs, in the atmosphere (Dunn, 2002).

360 362 362 363 363 365 365 367 367 370 372 373 373

The Kyoto protocol proposed that C reduction could take place by decreasing fossil fuel emissions, or by accumulating C in vegetation and in soils of terrestrial ecosystems. In various ecosystems, soil organic C (SOC) represents the largest reservoir of C that interacts with the atmosphere whereas vegetation stores considerably less C (Table 1). As such, fluxes between SOC and the atmosphere can either be positive (sequestration of CO2 ) or negative (emission of CO2 ) depending upon management practices of the terrestrial ecosystem. For example, SOC in agroecosytems, if managed sustainably, plays a critical role in the long-term storage of C in both tropical and temperate biomes. Canada’s potential for reducing atmospheric CO2 from agricultural practices is related to increasing and protecting soil organic matter (SOM). Under the Kyoto protocol, which Canada has ratified, Canada can now count these agricultural soil sinks as part of its commitment towards its emissions reduction target. Costa Rica has also played an important role in this process by being one of the first nations to implement a C trading system through the sustainable management and protection of existing forest and forest plantations, and the reforestation of previously deforested areas. One option to help address deforestation in tropical latitudes and create a C sink is to encourage the establishment of agroforestry systems. For example, Dixon (1995) estimated that for each hectare of sustainable agroforestry production, up to 5 ha of deforestation

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

361

Table 1 Global estimates of land area and C stocks in plant matter and soil ecosystems of the world (adapted from Amthor and Huston, 1998) Ecosystem

Area (×1012 m2 )

Plant C (g m−2 )

Soil Ca (g m−2 )

Total C (g m−2 )

Tundra Boreal forest Temperate forest Tropical forest Northern peatland Wetlands Temperate grassland Tropical savanna Chaparral Extreme desert, desert, semi-desert, scrub Lakes and streams Cultivated and permanent crop Temperate woodland Human area

9.5 9.0 7.5 14.8 3.4 2.8 12.5 22.5 2.5 30.0 2.0 14.8 2.0 2.0

630 2445 12270 16500 0 4300 720 2930 3200 365 10 200 8000 500

12750 15000 12000 8300 133800 72000 23600 11700 12000 10500 0 7900 12000 5000

13380 17445 24270 24800 133800 76300 24320 14630 15200 10865 10 8100 20000 5500

a

Soil C values are for the top 1 m, except for peatlands where they account for the total depth.

could be prevented. Sánchez (2000) stated that converting tropical forests into different types of agroforests results in a smaller loss of C compared to the conversion to croplands or pasture. In this paper, agroforestry is defined as the deliberate integration of woody species with agricultural crops and/or pastures on the same land-unit resulting in the integration of economical and ecological interactions between components (Lundgren, 1982; Nair, 1993; Young, 1997). Agroforestry also provides a sustainable alternative to shifting agriculture and to single crop systems because of its potential to help restore degraded or marginal soils (Dixon et al., 1993; Dixon, 1995; Young, 1997). For example, if agroforestry is established immediately after slash and burn agriculture, 35% of the original forest C stock can be regained (Sánchez, 2000). Agroforestry systems improve soil quality, through organic inputs from crop residues and tree litter, resulting in the maintenance or increase of SOM (Young, 1997). Higher levels of SOM improve crop yield and also stabilize soil C. The tree component in agroforestry systems can also act as a long-term C sink, depending upon end-use. For example, one-half to two-thirds of raw wood from forests is turned into wood products, representing an annual gross C flux of 0.3 Gt C year−1 (Watson et al., 2000). Carbon storage in timber products used for housing, particleboard and paper are examples of long-term C sinks derived from wood products. However, it should be noted that the net

accumulation of C in products is dependent upon the average lifetime of the product. The potential role of agroforestry systems to act as a C sink and to be integrated into a global C trading system has been overlooked. Therefore, integrating agroforestry systems into C sequestration and C trading projects may help to meet the CO2 emissions target proposed at Kyoto while at the same time maintaining sustainable agricultural production and preventing further deforestation in tropical latitudes. To date research in temperate and tropical agroforestry systems has focused on their efficacy in soil and water conservation, crop and pasture productivity, moisture and light competition, nutrient cycling, and changes in soil physical and chemical properties. The importance of agroforestry systems in CO2 mitigation is becoming more widely recognized. However, a knowledge gap exists on changes in SOC pools when converting from forests or sole cropped systems to agroforestry, on the long-term dynamics of SOC stabilization, on the rate of C accumulation in the tree component, and on the rate of C input from tree and crop residues and their contribution to stabilizing the soil C pool. The objective of this paper is to review and discuss C inputs derived from tree and crop residues, and to assess changes in the soil C pool as a result of agroforestry management practices in tropical and temperate biomes. The potential of agroforestry systems to sequester C in trees and/or in the soil will also be

362

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

discussed and contrasted between temperate and tropical areas. Information presented in this paper will help to address the emerging role of agroforestry systems in temperate and tropical biomes and their potential to sequester C.

2. Historical perspectives on agroforestry The cultivation of trees with agricultural crops dates to the beginning of plant and animal domestication (Smith, 1929; King, 1987; Williams et al., 1997). Since then, a variety of agroforestry systems have been developed in Asia, Africa, Europe, and parts of North and South America. These early agroforestry practices, like modern agroforestry systems, had a strong focus on sustainable crop production and soil conservation. For example, in Middle-Age Europe, degraded forest stands were clear-cut and seeded with crops. The slash was burned and crops were cultivated for varying time periods before new trees were planted again. Integrating apple orchards with sheep pasture or integrating timber or nut trees with cereal crops was also a common agroforestry practice in Europe (Gordon et al., 1997). In the tropics, farmers imitated vertical forest structures by planting a variety of crops with different growth habits, resulting in a high species diversity on a mall land area (Wilken, 1977; Kass and Somarriba, 1999). This system not only provided a diversity of crops to the farmer but it also protected the soil from erosion by reducing the impact from raindrops, and litter from trees provided organic material to sustain soil nutrient levels. Research on agroforestry systems did not begin until the mid 1970s. In response to increasing environmental degradation, the International Development Research Centre (IDRC) in Canada concluded that priority should be given to systems combining trees and crops to optimize sustainable land-use in areas with high population pressures (King, 1987). This led to a publication titled Trees, Food and People—Land Management in the Tropics (Bene et al., 1977), where the term ‘agroforestry’ was first used. This publication also illustrated that growing tees and agricultural crops on the same land area could help to conserve soil and improve crop productivity.

As a result, the International Council for Research in Agroforestry (ICRAF) was established in 1977 in Nairobi, Kenya. Since the establishment of ICRAF, agroforestry has been promoted as a sustainable landuse management system in both tropical and temperate latitudes. Modern experimental work in agroforestry began in the late 1970s including the first experiment on hedgerow intercropping (alley cropping) in Ibadan, Nigeria. Studies on nutrient cycling, using perennial crop combinations in Central America, and studies on the effectiveness of contour hedgerows on erosion control were also addressed (Young, 1997). Agroforestry systems in tropical regions have a different raison d’etre (reason for being) than those of temperate latitudes. In the tropics, agroforestry land management practices maintain landowner selfsustenance (Huxley, 1999), whereas in temperate latitudes the focus is on resource management policies, farming technology, labor costs and real estate values (Williams et al., 1997). However, in both biomes, trees are viewed as an integral part of agroforestry with the potential to restore degraded lands, to maintain soil fertility, and more recently to sequester C for mitigating atmospheric CO2 emissions.

3. Tropical and temperate agroforestry practices: examples from Costa Rica and Canada In Costa Rica problems with soil erosion and nutrient depletion due to single crop cultivation have persisted since traditional land management practices were abandoned over four decades ago. Although land management practices, where trees and agricultural crops are integrated on the same land area have been reported since the early 20th century (Cook, 1901), actual agroforestry research did not begin until the early 1980s. For example, the first alley cropping system was established in 1982 where maize (Zea mays L.), beans (Phaseolus sp.) and cassava (Manihot esculenta L.) were integrated with trees such as Erythrina poeppigiana Walp. (O.F.) Cook and Gliricidia sepium (Jacq.) Walp. This work was established in order to evaluate crop productivity, and the ability to maintain soil fertility without external inputs (Kass et al., 1992). Agroforestry is also practiced across Canada, and like in Costa Rica, there is a large variation in the type

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

of systems used and the adoption rate by local farming communities. There is still disagreement with respect to the definition of agroforestry; for example, in central Saskatchewan, the planting of fast-growing trees at high densities on marginal agricultural land is considered an agroforestry practice, although this might be referred to as aforestation in other areas of Canada. Southern Ontario remains one of the few geographical regions in Canada where all types of agroforestry, as defined by the Association for Temperate Agroforestry (AFTA), including intercropping (a type of alley cropping), silvopasture, integrated riparian systems, windbreaks and forest farming, are practiced (AFTA, 1994). Little difference exists between the major agroforestry systems in temperate and tropical biomes. For example, Gordon et al. (1997) describe a silvoarable system where timber trees are incorporated with annual crops in the temperate zone. This system is similar to the coffee (Coffea arabica L.)–E. poeppigiana–Cordia alliodora Ruiz (Lopez et Pavon) Cham. system in Costa Rica. In this system, coffee is the agricultural crop, and C. alliodora is the timber crop, whereas E. poeppigiana is a service tree. Small-scale forest farming of ginseng (Panax quinquefolius L.) is typical in some areas of North America, and is comparable to tropical homegardens (Gordon et al., 1997).

4. Carbon dynamics in tropical and temperate agroforestry systems It is projected that the area under agroforestry will increase substantially in order to address the growing demand for agricultural land in nations with rapidly rising populations, and rapidly decreasing resources. With more countries committing themselves to the Kyoto protocol, agroforestry systems are also becoming an economic incentive when integrated into a C trading system. Expansion of agroforestry systems therefore may impact the global C flux and the long-term C storage in terrestrial ecosystems (Dixon, 1995). The amount of C sequestered in agroforestry systems mostly depends upon the type of agroforest utilized. For example, shifting cultivation, pasture maintenance by burning, rice paddy cultivation, application of inorganic fertilizers, and animal production in silvopastoral systems,

363

are practices that increase GHG emissions to the atmosphere (Dixon, 1995). However, factors that influence the storage of C in agroforestry include system management (i.e. conservation tillage), use of groundcovers, fallowing, system age and design (i.e. tree densities), and tree species utilized. Fig. 1 illustrates C stocks and pools in 4-, 10- and 19-year-old alley cropping systems in Costa Rica compared to a 12-year-old hybrid poplar alley cropping system in southern Canada. Comparing these systems, C stocks and pools vary between tree age but not do not vary significantly between temperate and tropical systems of similar age. 4.1. Carbon input from aboveground tree components In tropical latitudes, alley cropping is an agroforestry practice where crops are grown between rows of trees that are regularly pruned. This agroforestry practice was initially developed to address issues related to soil fertility in subhumid and humid latitudes (Young, 1997). In this system, trees are completely shoot pruned (removing all branches and leaves) or partially pruned at regular intervals of 4–6 month. The organic material is spread on the soil surface as mulch, providing nutrients for the growing crops. Pruning productivity in alley cropping systems, and therefore the amount of C returned to the soil, ranges from 0.3 to 4.6 Mg C ha−1 (Table 2). Differences in climate, soil type, between and within tree species variation, and system management result in variable tree productivity (Kang and Wilson, 1987; Kass et al., 1992; Haggar et al., 1993; Mazzarino et al., 1993; Kang, 1997; Schroth et al., 2002). For example, management factors such as pruning frequency affect the nodulation efficiency in N2 -fixing species and hence overall tree productivity (Chesney, 2000; Chesney and Nygren, 2002). Tree productivity in Costa Rican alley cropping systems have also shown variable C input from prunings as a function of tree age and species (Table 2). For example, Nygren (1995) showed that C input from prunings of E. poeppigiana clones in Costa Rica ranged from 2.3 to 5.2 Mg C ha−1 year−1 at a tree density of 625 trees ha−1 . Oelbermann et al. (2003a) determined that C input from E. poeppigiana prunings varied with tree age, ranging

364

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377 200

Tree Leaves and Branches Tree Trunk

180

Tree Roots SOC

(a)

160

Carbon Pool (Mg C ha-1)

140 120 100 80 60 40 20 0 4-year

10-year

19-year

12-year

Fig. 1. Tree C stocks and SOC pools (to 40 cm) in 4-, 10-, and 19-year-old E. poeppigiana tropical alley cropping systems in Costa Rica and in a 12-year-old temperate hybrid poplar alley cropping system in southern Canada [data from Oelbermann, 2002; data for 19-year-old tree roots (a) from Nabuurs et al., 2001].

from 4.0 Mg C ha−1 year−1 in 19-year-old trees (555 trees ha−1 ) to 1.4 Mg C ha−1 year−1 in 10-yearold trees (833 trees ha−1 ). Alley cropping systems in the temperate zone are defined in the same way as those in the tropics, but their management is different. In temperate systems, trees are pruned more infrequently (every 2–5 years)

compared to the tropics. Due to their high lignin content and slow decomposition rates under temperate conditions, prunings are not applied to the cropped portion of the agroforest. Instead, prunings may be taken offsite, or chipped and spread within the tree row; a portion the agroforest that is not seeded with cash crops.

Table 2 Annual aboveground C inputs (Mg C ha−1 year−1 ) from multipurpose agroforestry tree prunings in tropical alley cropping systems Source

Location

Tree species

Agroforestry system

Soil type (FAO)

Age (years)

C input (Mg C ha−1 year−1 )

Oelbermann et al. (2003a) Oelbermann et al. (2003a) Oelbermann et al. (2003a) Oelbermann (2002) Oelbermann (2002) Oelbermann et al. (2003d) Adesina et al. (1999) Kang (1997) Kang (1997) Schroth and Lehmann (1995) Szott (1987)

Costa Rica Costa Rica Costa Rica Costa Rica Costa Rica Costa Rica Nigeria Nigeria Nigeria Central Togo Peru

E. poeppigiana E. poeppigiana E. poeppigiana G. sepium G. sepium G. sepium G. sepium G. sepium L. leucocephala G. sepium E. poeppigiana

Alley Alley Alley Alley Alley Alley Alley Alley Alley Alley Alley

Eutric Cambisol Eutric Cambisol Eutric Cambisol Eutric Cambisol Eutric Cambisol Eutric Cambisol Gleyic Solonetz Planosol Planosol Ferric Acrisol Planosol

4 10 19 4 10 19 7 6 6 4 NA

1.0 1.4 4.0 0.3 0.6 4.6 4.6 2.9a 4.2 0.8a 1.6a

a

crop crop crop crop crop crop crop crop crop crop crop

Carbon inputs were calculated from biomass inputs using a multiplication factor of 0.5.

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

Litterfall provides the largest flux of C from the tree component in a temperate alley cropping system. Carbon input from autumnal litterfall typically decreases exponentially with distance from the tree row. For example, a 12-year-old hybrid poplar alley crop in southern Canada, on an Albic Luvisol, contributed 0.95 Mg C ha−1 year−1 within 1 m of the tree row compared to 0.38 Mg C ha−1 year−1 at a 6.0 m distance (Oelbermann, 2002). Zhang (1999) reported litterfall to be 0.63 Mg C ha−1 year−1 in a 10-year-old hybrid poplar alley cropping system in southern Canada. Whereas Thevathasan and Gordon (1997) determined that leaves contributed 1.6 Mg C ha−1 year−1 by collecting all leaves from 7-year-old hybrid poplar at a stand density of 111 trees ha−1 . Litterfall productivity of alley crops in southern Canada has also shown to be increasing with age of the system. For example, Thevathasan and Gordon (1997) determined that C input from litterfall of 6-year-old hybrid poplar was 1.2 Mg C ha−1 year−1 and increased to 1.6 Mg C ha−1 year−1 in the following year. With increasing tree age, it is expected that C inputs to the soil C pool with also increase. This may also affect the distribution of C inputs within the cropped portion of the agroforest. As trees age, their canopies will eventually close over the cropped alley resulting in a more evenly distributed C input from litterfall. However, such a change in light conditions may lead to changing the type of crop grown. 4.2. The role of tree roots in agroforestry systems Roots in agroforestry systems comprise up to 30% of total tree biomass, and therefore they could play a significant role in increasing the belowground C pool (Young, 1997). Although the importance of roots in agroforestry research has been acknowledged, quantification of belowground biomass is difficult (Huxley, 1999), and most agroforestry research involving roots only began in the 1990s (Schroth, 1995; Atkinson, 1996; van Noordwijk et al., 1996). To date root research in agroforestry systems has focused on competitive interactions between trees and crops for moisture and nutrients (van Noordwijk et al., 1996; Jose et al., 2000; Lose et al., 2003). More recently work by Chesney and Nygren (2002) in Costa Rica determined the relationship between pruning frequency and fine root biomass and turnover, and nod-

365

ule formation in E. poeppigiana. However, only a few studies have quantified fine and coarse root biomass (Schroth, 1999; Dhyani and Tripathi, 2000; Livesley et al., 2000; Rowe et al., 2001) and root C (Schroth et al., 1996; Torquebiau and Kwesiga, 1996; Alegre et al., 2000) in tropical agroforestry systems. The studies cited above do not discuss the potential of roots to sequester C in agroforestry systems. However, Oelbermann et al. (2003b) determined that E. poeppigiana roots in 4- and 10-year-old alley cropping systems had an annual coarse root increment of 0.4 Mg C ha−1 year−1 compared to 0.2 Mg C ha−1 year−1 for the 4-year-old trees. The authors speculate that as trees age, their ability to allocate C to the root system will also increase, representing a significant long-term C sink in a tropical alley cropping system. Little data is available on roots in temperate agroforestry systems. Some information exists on fine root biomass distribution of black walnut (Juglans nigra L.) and northern red oak (Quercus rubra L.) in a southern Indiana alley crop system (Jose et al., 2001). No data is currently published on coarse root biomass in temperate alley cropping systems and its potential to sequester C. However, Thevathasan (2002, pers. com.) determined coarse root biomass of 12-year-old hybrid poplar to be 2.2 Mg C ha−1 , by completely exposing the root system within the soil profile. This would result in an annual C sequestration potential of 0.2 Mg C ha−1 year−1 at a stand density of 111 trees ha−1 . 4.3. Input from crop residues A large amount of C can be accumulated and returned to the SOC pool by cultivated plants, especially if only a portion of these plants are used as food or livestock feed. When crop residues are left on the soil surface, their nutrients are gradually released through mineralization by soil microorganisms. These nutrients are a source of food for soil biota, which drive the C and nutrient cycles. Therefore, the annual C input to the soil from crop residues plays a significant role in maintaining levels of SOC in agroecosystems and could help to mitigate atmospheric CO2 emissions. Sources of aboveground C from crop residues usually constitute all plant material except the harvestable grain or materials removed

366

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

Table 3 Carbon input (g C m−2 ) from maize and beans in a 10- and 19-year-old E. poeppigiana alley crop and sole crop in Costa Rica Maize

10-Year 10-Year 19-Year 19-Year

sole crop alley Crop sole crop alley crop

Total (g C m−2 )

Beans

Shoot

Root

Total (g C m−2 )

Shoot

Root

Total (g C m−2 year−1 )

42 201 115 158

7 29 18 20

49 230 121 178

27 62 19 28

7 9 7 7

34 71 26 35

(8) (23) (39) (20)

(2) (4) (6) (2)

(3) (9) (4) (3)

(1) (2) (2) (2)

83 301 147 213

Standard errors are given in parenthesis (Oelbermann et al., 2003a).

for livestock feed. Additional sources of annual C inputs from crops come from roots that remain in the soil after harvest, and the root C produced during the growing season, which is derived from root turnover and root exudates. Therefore, an evaluation of crop productivity and residue returned to the soil is timely because of its potential influence on the global C balance and the SOC pool (Buyanovsky and Wagner, 1986; Bolinder et al., 1997). However, data on crop productivity and C input in agroforestry systems, particularly alley crops, is scant. Instead, most agroforestry research has focused on the economic yield of cultivated plants. Competition for resources between trees and growing crops is a major factor leading to a reduction in crop productivity in alley crops compared to sole crops. For example, some studies determined that crop yields are lower next to the tree row compared to the center of the cropped alley in tropical and temperate systems (Yamoah et al., 1986; Huxley et al., 1989; Rao et al., 1991; Chamshama et al., 1998; Miller and Pallardy, 2001). On the contrary, alley cropping also has the potential to improve soil conditions. For example, a rise in soil nutrient status is commonly observed in alley crops with a high input of organic material, which in turn increases crop productivity (Mahboubi et al., 1997). A higher grain yield and crop productivity was reported by Kang and Wilson (1987) and Yamoah et al. (1986)

working on a Gleyic Solonetz in Nigeria as a result of adding G. sepium, L. leucocephala and Flemingia contesta Aiton f. tree prunings to the soil surface in a maize alley crop compared to a sole crop. Studies from Costa Rica reported an increase in aboveground maize and bean productivity after 6 years (Sánchez, 1989; Kass et al., 1992), and after 10 and 19 years of alley cropping compared to their respective sole crops (Table 3). Haggar et al. (1993) found that 7 years of alley cropping in Costa Rica with E. poeppigiana and G. sepium resulted in higher maize biomass and maize N content in alley crops than sole crops by 120 and 180%, respectively. They concluded that the major effect of N input from mulch, which lead to an increase in N availability, was through increasing the size of the readily mineralizable N pool over the medium term of 7–8 years. In southern Canada, C input from crop residues do not differ greatly from tropical systems, despite two crop rotations per annum in the tropics (Table 4). However, crop productivity may be adversely affected as trees age in southern Canadian alley crops. Ntayombya and Gordon (1995) found that alley cropping on a Grey Brown Luvisol did not affect barley (Hordeum vulgare L.) productivity and N nutrition when grown with black locust (Robinia pseudoacacia L.) during the first year of alley cropping. However, crop productivity declined significantly in the following years when trees grew bigger.

Table 4 Carbon input (mean of two seasons) from crop residues in a 12-year-old hybrid poplar alley cropping system in southern Canada

Maize Soybeans Wheat

Shoots (g C m−2 year−1 )

Roots (g C m−2 year−1 )

Total (g C m−2 year−1 )

177 (11) 72 (14) 107 (15)

29 (11) 8 (2) 15 (4)

206 80 122

Crops are grown on a 3-year rotation in the hybrid poplar alley crop. Standard errors are given in parentheses (Oelbermann, 2002).

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

A follow-up study by Thevathasan and Gordon (1995), using a hybrid poplar–barley system (replicated potted experiment), did not show a decline in crop productivity and grain yield. They suggested that competition for moisture and nutrients was not the factor that affected crop productivity, because trees were exploiting resources from lower soil horizons than the crops. In temperate and tropical alley cropping systems, space available to trees and crops affects productivity of both. Imo and Timmer (1999) found that maize productivity decreased when the width between L. leucocephala alleys was 2 m, but was not affected when the alley width was increased to 8 m. It was also observed that high crop productivity coincided with low pruning biomass as a result of resource competition (Seiter et al., 1999). These results suggest that tree spacing is an important consideration when designing alley cropping systems in the temperate zone. Jose et al. (2000) propose that wider spacing between tree rows can maintain crop productivity similar those of sole crops. 4.4. Decomposition and stabilization of organic matter The organic C stock present in the soil represents a dynamic balance between the input of dead plant material and the loss from decomposition (Fig. 2). Carbon is stored in diverse forms with a wide range of mean residence times (Jenkinson and Rayner, 1977; Schimel, 1994; Saggar et al., 1994; Torn et al., 1997). For example, labile C such as fresh litter or nonwoody material is quickly decomposed with a mean residence time of 3–4 years. Whereas woody materials become part of the active soil pool which may persist as long as 1000 years or more (Parton et al., 1987). This persistent sock of C is stable because it is chemically inert or physically protected (Parton et al., 1987; Paul et al., 1997). Therefore, C incorporated into more stable pools can produce slow, but long-term increases of soil C. Alley cropping systems have the ability to build stable soil C pools through their high return of organic material from tree prunings in tropical areas and litterfall in the temperate zone. The quantity of C and nutrients added to the soil from tree prunings or litterfall is directly dependent on tree productivity, tree species,

367

system management (i.e. frequency of pruning, tree density), and site-specific edaphic and climatic conditions (Nair, 1993). The quality of organic inputs is also a key factor in controlling the rate of decomposition and nutrient release (Kwabiah et al., 1999, 2001). The rate of decomposition from tree prunings and crop residues in tropical alley cropping systems is rapid when tree leaves or fresh litter is added to the soil surface. In these systems, the release of nutrients of added organic material is synchronized with crop seeding. However, leaves are a labile source of C in these systems and therefore emit CO2 to the atmosphere instead of its long-term storage in the soil (Tian et al., 1992; Lehmann et al., 1995; Henrot and Brussaard, 1997; Mugendi and Nair, 1997; Vanlauwe et al., 1997; Mugendi et al., 1999; Isaac et al., 2000; Vanlauwe et al., 2001). However, typical mulch systems in tropical alley crops also incorporate woody material from tree prunings, which increases the mean residence time of C within the soil due to its slower rate of decomposition. Incroporating woody material into mulch would help to reduce CO2 emissions per unit of soil C per unit of time. In the temperate zone, decomposition of litterfall is slower because of climatic conditions, which increases the amount of C entering the stable C pool. To date, limited data are available on decomposition of mulch and litterfall in tropical and temperate alley cropping systems and its influence on the soil C pool. For example, Quinlan (1984), Haggar et al. (1993) and Oelbermann et al. (2003c) determined decomposition of tree prunings and crop residues in Costa Rican alley cropping systems, however there is a lack of this information in temperate agroforestry systems (i.e. Jose et al., 2000; Abohassan, 2004). 4.5. Soil organic carbon pools Soil is an irreplaceable resource as its formation takes place over centuries (Miller and Gardiner, 2001), although remediation strategies can promote soil development within decades (Hudson, 1995). Our current approach to soil conservation is based on tolerable levels of soil loss ranging from 4.5 to 11.2 Mg ha−1 year−1 (Hudson, 1995). The global SOC pool contains 1500 Pg C and is two to three times greater than the amount of C stored in vegetation (Eswaran et al., 1993). However, cultivation

368

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

Fig. 2. Model of soil organic C dynamics and stabilization. In this model, soil is 20% clay, water pore volume >0.4, temperature is 12 ◦ C, annual crops, and conventional tillage (adapted from Balesdent et al., 2000).

of virgin soil results in large losses of SOC (Paustian et al., 2000; Solomon et al., 2002). For example, when tropical forests are cultivated, SOC losses to a 1 m depth range from 15 to 40% within 2–3 years (Ingram and Fernandes, 2001), reducing soil fertility and crop productivity. Other factors that have been classified as immediate causes of a decline in SOC include residue removal, soil erosion, intensive tillage, and bare fallowing (Lal and Kimble, 2000; Paustian et al., 2000).

Factors leading to soil loss and degradation in the temperate zone parallel those in the tropics. Management strategies to increase SOC should be directed towards increasing residue inputs and/or decreasing decomposition rates (Batjes and Sombroek, 1997; Lal and Kimble, 2000), which can be achieved by crop rotation, the use of cover crops and vegetative fallows (Rosenzweig and Hillel, 2000). For example, Izaurralde et al. (2001) estimate that globally

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

369

Table 5 Biologic potential of annual C sequestration in aboveground biomass using different land-use practices in temperate and tropical biomes (adapted from Dixon and Turner, 1991) Land management practice

Tropical biomes (×109 Mg C year−1 )

Temperate Biomes (×109 Mg C year−1 )

Forestation Agroforestry Land rehabilitation Conservation agriculture End deforestation and desertification Total

1.3 2.1 0.1 2.4 2.8 8.7

0.9 1.9 0.1 1.0 1.0 4.9

over 2 billion ha of degraded land exists, of which 1.5 billion ha are located within tropical latitudes. Restoration of these degraded lands through aforestation, agroforestry, land rehabilitation, elimination of deforestation and desertification has the biotic potential to sequester 8.7 × 109 Mg C year−1 in the tropical and 4.9 × 109 Mg C year−1 in the temperate aboveground C pools (Dixon and Turner, 1991). Of this pool, 2.1 × 109 and 1.9 × 109 Mg C year−1 could be sequestered in aboveground components of the terrestrial biosphere and therefore help to conserve existing soil organic C (Table 5). Therefore, a common goal in both latitudes should be the conservation of biophysical resources, which can be accomplished by sustainable land management practices such as agroforestry within a time frame of decades. For example, growing trees in association with agricultural crops is more efficient compared to single crop systems in temperate and tropical latitudes (Huxley, 1999). This is because woody species have deeper rooting systems, and can capture nutrients from lower soil horizons (Nair, 1993). Additionally, organic matter input from tree prunings and litterfall provide a source of nutrients; maintain SOC levels, and balance soil moisture and temperature, which also affect soil microbiological activity. For example, Kaur et al. (2000) noted that soil microbial C was 42% greater in an agroforestry system compared to a sole crop in India. In Costa Rica, Mazzarino et al. (1993) found that soil C and N, and microbial C and N, were greater in alley crops with E. poeppigiana and G. sepium than in a sole crop. Therefore, the perennial nature of agroforestry systems results in a continuous flux of C and nutrients, preventing the uncoupling of soil C and nutrients, which are released during plant uptake.

The importance of organic matter input from tree prunings and litterfall, to help maintain or increase the SOC pool, has been demonstrated by several studies in tropical and temperate agroforestry systems. For example, Dulormne et al. (2000) reported a 15% increase in SOC to a 20 cm depth after 10 years of silvopasture with G. sepium in the French Antilles (Table 6). In Nigeria, Kang (1997) found that SOC increased by 37% when maize was grown with Leucaena leucocephala (Lam.) De Wit in an alley crop compared to a sole crop. In Costa Rica, inputs of organic material from tree prunings and litterfall in a Theobroma cacao L.–E. poeppigiana–shade tree system increased levels of SOC from 115 to 140 Mg C ha−1 (to a 45 cm depth) over a 9-year-period (Fassbender, 1998). Although little information exists on increases in SOC in Costa Rican alley cropping systems, Oelbermann (2002) and Oelbermann et al. (2003d) determined that after 19 years the SOC pool was significantly greater in a E. poeppigiana and G. sepium hedgerow system compared to a sole crop. However, in a 10-year-old alley crop, using the same tree species as the 19-year-old system, no differences in the SOC pool size was determined between the alley and sole crop (Table 6). This indifference in the SOC pool between the alley and sole crop is likely due to the age of the system. Young (1997) suggested that in tropical environments, at least a 10-year-period of alley cropping is necessary before any changes in the levels of SOC can be detected. Therefore, alley cropping systems could function as a C sink in the soil component after a decade of establishment. However, Nabuurs et al. (2001) determined that the tree component, including the root system, became a C sink 4 years after the establishment of a C. alliodora–E.

370

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

Table 6 Soil organic carbon (SOC [mg g−1 ]) and soil bulk density (BD [Mg m−3 ]) to a 20 cm depth in tropical agroforestry systems Source

Oelbermann (2002) Oelbermann (2002) Oelbermann (2002) Oelbermann (2002) Schroth et al. (2002) Dulormne et al. (2000) Tornquist et al. (1999) Aihou et al. (1998) Chander et al. (1998) Fassbender (1998) Kang (1997) Mazzarino et al. (1993) Mazzarino et al. (1993) Haggar (1990) Haggar (1990) a b

Location

Costa Rica Costa Rica Costa Rica Costa Rica Brazil French Costa Rica B´enin India Costa Rica Nigeria Costa Rica Costa Rica Costa Rica Costa Rica

Tree species

E. poeppigiana E. poeppigiana G. sepium G. sepium Multistrata system Antilles G. sepium Vochysia ferruginea G. sepium Dalbergia sissco E. poeppigiana L. leucocephala E. poeppigiana G. sepium E. poeppigiana G. sepium

Soil type (FAO)

Eutric Cambisol Eutric Cambisol Eutric Cambisol Eutric Cambisol Xanthic Ferrasol Vertisol Acrisol Rhodic Ferrasol Haplic Chernozem Eutric Cambisol Planosol Eutric Cambisol Eutric Cambisol Eutric Cambisol Eutric Cambisol

Age (years)

10 19 10 19 7 10 10 10 13 9 5 10 10 6 6

Agroforestry

Sole crop

SOC

BD

SOC

BD

19.0 29.0 29.9 32.0 22.2a 28.4 36.5b 47.0 6.2b 27.3b 9.4b 27.8a 27.5a 18.5 14.8

1.2 1.2 1.2 1.1 0.8 1.7 1.0 – – 1.2 – 1.2 1.2 1.0 1.0

18.7 24.1 18.7 24.0 19.0a – – 3.90 6.5b – 5.9b 23.1a 23.1a 13.6 13.6

1.1 1.2 1.1 1.2 1.0 – – – – – – 1.2 1.2 1.0 1.0

0–10 cm depth. 0–15 cm depth.

poeppigiana–coffee shade tree system in Costa Rica on a Eutric Cambisol. In the temperate zone, increases in the SOC pool have also been reported as a result of alley cropping. For example, SOC increased by 4 and 7%, respectively, in an alley cropping system in western Oregon with Alnus rubra Bong. and R. pseudoacacia L. compared to a sole crop. A parallel study conducted by Oelbermann (2002) comparing SOC pools in a 19-year-old tropical and 12-year-old temperate alley crops in Costa Rica and Canada found that the SOC pools were similar in the two biomes, although inputs were lower in the temperate system. This is because the turnover rate of organic material is faster in the tropical system. Given the slower rate of SOC turnover and C stabilization in the temperate zone, it may be that under these climatic conditions a period much greater than 10 years is needed to observe any changes in the SOC pool, yet the tree component can likely function as a C sink as soon as the system is established.

5. Linking agroforestry to carbon sequestration and carbon trading It is estimated that the atmospheric CO2 concentration will double by the mid to late 21st cen-

tury (Winjum et al., 1993; Houghton et al., 1993; Hengeveld, 2000). This increase of atmospheric CO2 could elevate global temperatures by 1.5–4.5 ◦ C, leading to large shifts in the production and distribution of vegetation (Dotto, 2000; William et al., 2000). These shifts will also affect agricultural crop productivity and increase the total land area of degraded soils (Cole et al., 1993; Dixon et al., 1993; Rosenzweig and Hillel, 2000). Agroforestry practices like alley cropping and silvopasture have the greatest potential for conserving and sequestering C because of the close interaction between crops, pasture, trees and soil (Nair, 1998). Therefore agroforestry systems have a direct near-term (decades or centuries) C storage capability in trees and soils, and have the potential to offset immediate greenhouse gas emissions associated with deforestation and shifting cultivation (Dixon, 1995; Nair and Nair, 2002). Sánchez (2000) noted that agroforestry systems could be superior to other land-uses at the global, regional, watershed, and farm level because they optimize tradeoffs between increased food production, poverty alleviation and environmental conservation. However, they can also be inferior to other land-use systems because of inappropriate technologies, and accompanying policies are not enabling farmers to establish the system as a C offset project (Sánchez, 2000).

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

The incorporation of trees on farms affects C stocks differently than in a typical cropland or forest land management system. Trees in agroforestry systems provide a tighter coupling of nutrients and maintenance of the SOC pool compared to sole cropped areas. But trees are also harvested more frequently in these systems compared to a typical forest management program (Watson et al., 2000). However, trees in agroforestry system can have rotation times similar to those of a forest management program. For example, in Costa Rica, E. poeppigiana was established as an alley cropping system over 20 years ago with very low tree mortality (Kass, 2003, pers. com.). Some coffee plantations with E. poeppigiana as a shade tree were established over 100 years ago, where trees have been pollarded biannually for the inclusion of prunings as mulch (Kass, 2003, pers. com.). Nabuurs et al. (2001) designed a model (CO2 Fix) to predict the C budget of forest ecosystems as a standard for the quantification of C sequestration in aforestation projects, and sustainable forest management, including agroforestry systems. They projected that a 5-year-old coffee plantation with E. poeppigiana and C. alliodora could sequester 5.3 Mg C ha−1 . The model projected that this agroforestry system would sequester C steadily, reaching 183.5 Mg C ha−1 in its 100th year. The model did not predict C sequestration beyond year 100. Watson et al. (2000) projected that C stocks for smallholder agroforestry systems in the tropics sequestered C ranging from 1.5 to 3.5 Mg C ha−1 year−1 with a tripling of C stocks in a 20-year-period to 70 Mg C ha−1 year−1 . With such potential of agroforestry systems to store C in trees and in the soil, these systems should be included in C sequestration and C trading projects. Carbon trading allows industries in developed countries to offset their C emissions by investing in reforestation and in clean energy projects in developing countries. For example, in 1997 Costa Rica established certified tradable offsets (CTOs) in order to save more than 3 × 105 ha of rainforest by selling carbon credits to northern European countries (Allen, 1998; Goodman, 1999). This was the first C trading project developed under the guidelines of the Kyoto protocol (Allen, 1998; Wright et al., 2000). These bonds are now sold at US$10.00 per Mg of

371

C conserved, providing landowners with an incentive for sustainable forest management (Goodman, 1999; Wright et al., 2000). If the Kyoto protocol is implemented and enforced with reaching the proposed mandatory C target in 2012, the price of C is projected to increase to over US$30 Mg−1 (Dayal, 2000). To date these credit services in Costa Rica have focused on private land reforestation, the management and protection of existing forests, and on plantation forests (MINAE, 2002). As a result, 3000 farmers have participated in Costa Rica’s carbon credit system conserving 150,000 ha of forest. In late 2002, the Executive Director of the Costa Rican National Agroforestry Commission recommended that agroforestry land management practices should be included in the C credit system (MINAE, 2002). The Costa Rican government has now budgeted $400,000 for the integration of agroforestry land management practices into the C trading system (Ibrahim, 2002, pers. com.). Carbon trading or C offset programs have also been initiated in the temperate zone. For example, Hoever et al. (2000) noted an existing market for C credits within the United States that is driven by utility companies, which initiated forestry projects specifically designed to offset C emissions. Landowners that could increase C storage through sustainable forest management were also able to sell C as credits. In early 1998, the Canadian firm Suncor Energy, Inc. purchased 100,000 Mg of C credits, with an option to purchase an additional 10 million Mg from Niagara Mohawk Power Corp (Dayal, 2000). Niagara Mohawk reduced green house gas emissions below 1990 levels through improvements in energy-use efficiency. This deal helped Suncor Energy achieve a reduction in emission targets while Niagara Mohawk obtained funding for new greenhouse gas project. However, lack of reliable estimates of how much area is currently under agroforestry production in different ecological zones is an obstacle. It is currently not known to which extent agroforestry practices could counter C emission from deforestation and their effectiveness in C trading. However, the International Panel on Climate Change (IPCC) estimates that the current worldwide area under agroforestry is 400 million ha, which results in a C gain of 0.72 Mg ha−1 year−1 (Watson et al., 2000). It is estimated that the potential C gain could increase to 26 × 106 Mg ha−1 year−1

372

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

by 2010 and to 45 × 106 Mg ha−1 year−1 by 2040 (Watson et al., 2000). As a result of intensive land management practices in both temperate (Rosenzweig and Hillel, 2000) and tropical (Schroeder, 1994) regions, a large area of degraded land has emerged that is suitable for initiating agroforestry land management practices. It has been estimated that globally, 630 million ha of unproductive cropland and grassland could be converted to agroforestry worldwide with a potential to sequester 391× 106 Mg C year−1 by 2010 and 586 × 106 Mg C year−1 2040 (Watson et al., 2000). The IPCC estimated that agroforestry can sequester C at a time-average rate of 0.2–3.1 Mg C ha−1 . In Costa Rica as much as 876,700 ha of land exist on which agroforestry systems could be established (Maldonado, 2002, pers. com.). Oelbermann (2002) showed that an E. poeppigiana alley cropping system in Costa Rica was able to increase SOC by 1.1 Mg C ha−1 year−1 . Using the potential land area available for agroforestry alley cropping systems in Costa Rica, a total of 9.6 × 105 Mg C year−1 could be sequestered in soil. The tree component of a 10-year-old E. poeppigiana alley cropping system, at a stand density of 833 trees ha−1 , sequestered C at a rate of 0.04 Mg C ha−1 year−1 in fine roots (Chesney, 2000) and 0.4 Mg C ha−1 year−1 in coarse roots to a 60 cm depth, 0.3 Mg C ha−1 year−1 in the trunk, and 1.4 Mg C ha−1 year−1 in branches and leaves (Oelbermann et al., 2003b). However, some of the C sequestered in leaves and branches is returned to the soil C pool as it decomposes as mulch on the soil surface and the remainder is emitted to the atmosphere as CO2 . This resulted in an annual increase of 0.6 Mg ha−1 year−1 in the soil C pool. Using the potential land area available for agroforestry systems in Costa Rica, a total of 6.5 × 105 Mg C year−1 could be sequestered in tree roots and trunks if alley cropping systems were established on this area. In the temperate zone, the C sequestration potential for agroforestry systems ranges from 15 to 198 Mg C ha−1 with a modal value of 34 Mg C ha−1 (Dixon et al., 1993). However, Garrett and McGraw (2000) acknowledge that reliable data of the area under alley cropping in North America is not available, but they suggest that more than 45 million ha of nonfederal cropland in the United States could be

available for alley cropping. Based on these data, in addition to the land available for silvopasture, windbreaks, and riparian buffers, Nair and Nair (2002) estimate that the total C sequestration potential of all agroforestry practices in the United States could amount to 90 × 106 Mg C year−1 . Canada is responsible for 2% of global CO2 emissions, and Canada’s commitment to the Kyoto protocol is to reduce CO2 emissions to 6% below 1990 levels by 2012, which would equal a reduction of 65 × 106 Mg of C (Dunn, 2002). Thevathasan (2003, pers. com.) suggests that 7 × 106 ha of Canada’s land area have been classified as marginal agricultural land, of which 40% is suitable for alley cropping. Thevathasan et al. (2000) estimated that a hybrid poplar (Populus deltoides × nigra DN-177) alley cropping system (wheat–soybean–maize rotation, on marginal soil) could sequester 0.8 and 0.2 Mg C ha−1 year−1 in aboveground and belowground components, respectively, at a stand density of 111 trees ha−1 . Therefore, if 40% of the marginal agricultural land is alley cropped with hybrid poplar, at 111 trees ha−1 , up to 2.1 × 106 and 5.0 × 105 Mg C year−1 could be sequestered in aboveground and belowground tree components. Over the next 8 years, Canada could sequester 16.8 × 106 and 4 × 106 Mg C in aboveground and belowground tree components, representing 32% of Canada’s contribution to its commitment to the Kyoto protocol.

6. Conclusions and future research needs Biophysical interactions in agroforestry systems including soil fertility management, soil improvement by trees, and tree-crop competition have been the main area of applied research to date. Although some work has focused on the design of agroforestry systems, there are still gaps in the literature on alley cropping design to maximize crop yields (Wojtokowski, 1998). The potential role of agroforestry systems to sequester C was addressed in an IPCC report (Watson et al., 2000), however data on C sequestration and the role of C trading in agroforestry systems are still in its infancy. For example, no information is currently available to design agroforestry system as a C sink while at the same time maintaining crop productivity.

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

Results from Oelbermann (2002) have shown that SOC can increase significantly in Costa Rican alley cropping systems over a 19-year-period using a minimum tree density of 555 trees ha−1 for E. poeppigiana and 3333 trees ha−1 for G. sepium. These alley cropping systems also maintained or improved grain yields compared to sole crops (Oelbermann, 2002). Results from Canada have shown that hybrid poplar alley cropping system with a tree density of 111 trees ha−1 could contribute significantly to reducing Canada’s current levels of CO2 emissions. Hybrid poplar alley cropping systems on marginal land in Canada could help to mitigate 32% of the required CO2 reduction level for Canada by 2012 in aboveground and belowground tree components. Future research should focus on the use of stable isotope 13 C in SOC studies, which helps to distinguish between soil C derived from C3 and C4 vegetation. This will generate a better insight into the dynamic changes and stabilization of SOC following conversion of forests or pasture to agroforests. Further emphasis should also be placed on integrating agroforestry systems into the C credit system, particularly in nonindustrialized nations, which would greatly benefit from the additional monetary influx.

Acknowledgements The authors sincerely thank the Natural Sciences and Research Council of Canada (NSERC) and the International Institute for Cooperation on Agriculture (IICA-Canada) for its support for research conducted in Costa Rica and Canada. Special thanks also to Drs. D.C.L. Kass, A.M. Schlönvoigt, and M. Ibrahim for valuable comments on tropical agroforestry systems, and to Alexis Pérez, Francisco Nuñez, Joaqu´ın Soto, and Marvin Saborio for technical support during field work in Costa Rica. References Abohassan, R.A., 2004. Carbon dynamics in a temperate agroforestry system in southern Ontario, Canada. M.Sc. Thesis, University of Guelph, Guelph, Canada, p. 118. Adesina, F.A., Siyanbola, W.O., Okelota, F.A., Pelemo, D.A., Momudu, S.A., Adegbulugbe, A.O., Ojo, L.O., 1999. Potential of agroforestry techniques in mitigating CO2 emissions in

373

Nigeria: some preliminary estimates. Glob. Ecol. Biogeogr. 8, 163–173. AFTA, 1994. Agroforestry: an integrated land-use management system for production and farmland conservation. The Agroforestry Component of the Resource Conservation Act Appraisal for the USDA-SCS. Aihou, K., Sanginga, N., Vanlauwe, B., Lyasse, Ol., Dields, J., Mercks, R., 1998. Alley cropping in the moist savanna of West Africa. 1. Restoration and maintenance of soil fertility on ‘terre de barre’ soils in Bénin Republic. Agrofor. Syst. 42, 213–277. Alegre, J., Arevalo, L., Ricse, A., 2000. Reservas de carbono con diferentes sistemas de uso de la tierra en dos sitios de la Amazonia Peruana. ICRAF-Peru. AgroFor2-L, p. 9. Allen, V., 1998. Costa Rica to save forests with carbon credits. Reuters, April 24, 1998. Amthor, J.S., Huston, M.I., 1998. Terrestrial Ecosystem Responses to Global Change: A Research Strategy, ORNL/TM-1998/27. Oak Ridge National Laboratory. Atkinson, D., 1996. Why study roots? Agrofor. Forum 7, 2–4. Balesdent, J., Arrouays, D., Gaillard, J., 2000. MORGANE: un modèle de simulation des réserves organiques des sols et de la dynamique du carbone des sols. Agronomie 20, 3–10. Batjes, N.H., Sombroek, W.G., 1997. Possibilities for carbon sequestration in tropical and subtropical soils. Glob. Change Biol. 3, 161–173. Bene, J.G., Beall, H.W., Cˆoté, A., 1977. Trees, Food and People—Land Management in the Tropics. IDRC, Ottawa, Canada. Bolinder, M.A., Angers, D.A., Dubuc, J.P., 1997. Estimating shoot to root ratios and annual carbon inputs in soils for cereal crops. Agric. Ecosyst. Environ. 63, 61–66. Buyanovsky, G.A., Wagner, G.H., 1986. Post-harvest residue input to cropland. Plant Soil 93, 57–65. Chamshama, S.A.O., Mugasha, A.G., Klivstad, A., Haveraaen, O., Maliondo, S.M.S., 1998. Growth and yield of maize alley cropped with Leucaena leucocephala and Faidherbia albida in Morogoro, Tanzania. Agfor. Syst. 40, 215–225. Chander, K., Goyal, S., Nandal, D.P., Kapoor, K.K., 1998. Soil organic matter, microbial biomass and enzyme activities in a tropical agroforestry system. Biol. Fertil. Soils 27, 168–172. Chesney, P.E.K., 2000. Pruning effects on roots of nitrogen fixing trees in the humid tropics. Ph.D. Thesis. CATIE, Turrialba, Costa Rica, p. 222. Chesney, P.E.K., Nygren, P., 2002. Fine root and nodule dynamics of Erythrina poeppigiana in alley cropping systems in Costa Rica. Agrofor. Syst. 56, 259–269. Cole, V.C., Lee, J., Sauerbeck, D., Stewart, B., 1993. Agricultural sources and sinks of carbon. Soil Water Air Pollut. 70, 111–122. Cook, O.F., 1901. Shade in Coffee Culture. US Department of Agriculture, Division of Botany, Washington, DC. Costa, M.H., Foley, J.A., 2000. Combined effects of deforestation and doubled atmospheric CO2 concentrations on the climate of Amazonia. J. Climate 13, 18–34. Dayal, P., 2000. Carbon trading and sequestration projects offer global warming solutions. Environ. Manage. 3, 15–24. Dhyani, S.K., Tripathi, R.S., 2000. Biomass and production of fine and coarse roots of trees under agrisilvicultural practices in north-east India. Agrofor. Syst. 50, 107–121.

374

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

Dixon, R.K., Turner, D.P., 1991. The global carbon cycle and climate change: Responses and feedback from below-ground systems. Environ. Pollut. 73, 245–261. Dixon, R.K., Winjum, J.K., Schroeder, P.E., 1993. Conservation and sequestration of carbon: the potential of forest and agroforestry management practices. Glob. Environ. Change 3, 159–173. Dixon, R.K., 1995. Agroforestry systems: sources and sinks of greenhouse gases? Agrofor. Syst. 31, 99–116. Dotto, L., 2000. Proof or consequences. Alternat. J. 26, 2–10. Dulormne, M., Sierra, J., Sophie, S.A., Solvar, F., 2000. Capacidad de secuestración del carbono y del nitrógeno en un sistema agroforestal a base de Gliricidia sepium en clima tropical sub-húmedo (Guadalupe, Antillas Francesas). Unité Agropédoclimatique, Institut National de la Recherche Agonomique, Domaine Duclos, Prise d”Eau, 97170 Tetit Bourg, Guadeloupe, France. Dunn, S., 2002. Reading the Weathervane: Climate Policy from Rio to Johannesburg, Worldwatch Paper No. 160. Worldwatch Institute, Washington, DC. Eswaran, H., van den Berg, E., Reich, P.F., 1993. Organic carbon in soils of the world. Soil Sci. Soc. Am. J. 57, 193–194. Fassbender, H.W., 1998. Long-term studies of soil fertility in cacao-shade trees agroforestry systems: results of 15 years of organic matter and nutrients research in Costa Rica. In: Schulte, A., Ruhujat, D. (Eds.), Soils of Tropical Forest Ecosystems: Characteristics, Ecology and Management. Springer Verlag, Berlin, pp. 150–158. Garrett, H.E., McGraw, R.L., 2000. Alley cropping practices. In: Garrett, H.E., Rietveld, W.J., Fisher, R.F. (Eds.), North American Agroforestry: An Integrated Science and Practice. American Society of Agronomy, Madison, WI, pp. 149–188. Goodman, A., 1999. Whispers from the vaults: carbon trading up and running—Costa Rica is an early player in carbon trading—sequestration of forest acreage is a salable asset. Tomorrow Glob. Environ. Bus. 8, 28. Gordon, A.M., Newman, S.M., Williams, P.A., 1997. Temperate agroforestry: an overview. In: Gordon, A.M., Newman, S.M. (Eds.), Temperate Agroforestry Systems. CAB International, Wallingford, UK, pp. 49–89. Haggar, J.P., 1990. Nitrogen and phosphorus dynamics of systems integrating trees and annual crops in the tropics. Ph.D. Thesis. St. John’s College, University of Cambridge, UK, p. 175. Haggar, J.P., Tanner, E.V.J., Beer, J.W., Kass, D.C.L., 1993. Nitrogen dynamics of tropical agroforestry and annual cropping systems. Soil Biol. Biochem. 25, 1363–1378. Hengeveld, H.G., 2000. The science: global temperature is on the rise and the dangers are real and significant. Alternatives 26, 11–12. Henrot, J., Brussaard, L., 1997. Determinants of Flemingia contesta and Dactyladenia barteri mulch decomposition in alley cropping systems in the humid tropics. Plant Soil 191, 101–107. Hoever, C.M., Birdsey, R.A., Heath, L.S., Stout, S.L., 2000. How to estimate carbon sequestration on small forest tracts. J. For. 98, 13–19. Houghton, R.A., Unruh, J.D., Lefebvre, P.A., 1993. Current land cover in the tropics and its potential for sequestering carbon. Glob. Biogeochem. Cycles 7, 305–320.

Hudson, N., 1995. Soil and Water Conservation. B.T. Batsford Ltd., London, UK. Huxley, P., 1999. Tropical Agroforestry. Blackwell Science, Oxford, UK. Huxley, P., Pinney, A., Akunda, E., Gatama, D., 1989. The tree/crop interface: a project designed to generate experimental methodology. Agrofor. Abstr. 2, 127–145. Imo, M., Timmer, V.R., 1999. Vector competition analysis of a Leucaena–maize alley cropping system in western Kenya. For. Ecol. Manage. 47, 1–14. Ingram, J.S.I., Fernandes, E.C.M., 2001. Managing carbon sequestration in soils: concepts and terminology. Agric. Ecosyst. Environ. 87, 111–117. Isaac, L., Wesley Wood, C., Shannon, D.A., 2000. Decomposition and nitrogen release of prunings from hedgerow species assessed for alleycropping in Haiti. Agron. J. 92, 501–511. Izaurralde, R.C., Rosenberg, N., Lal, R., 2001. Mitigation of climatic change by soil carbon sequestration: issues of science, monitoring and degraded lands. Adv. Agron. 70, 1–75. Jenkinson, D.S., Rayner, J.H., 1977. The turnover of soil organic matter in some of the Rothamsted classical experiments. Soil Sci. 123, 298–305. Jose, S., Gillespie, A.R., Seifert, J.R., Mengel, D.B., Pope, P.E., 2000. Defining competition vectors in a temperate alley cropping system in the midwestern USA. 3. Competition for nitrogen and litter decomposition dynamics. Agrofor. Syst. 48, 61–79. Jose, S., Gillespie, A.R., Seifert, J.R., Pope, P.E., 2001. Comparison of minirhizotron and soil core methods for quantifying root biomass in a temperate alley cropping system. Agrofor. Syst. 52, 161–168. Kang, B.T., Wilson, G.F., 1987. The development of alley cropping as a promising agroforestry technology. In: Steppler, H.A., Nair, P.K.R. (Eds.), Agroforestry: A Decade of Development. ICRAF, Nairobi, pp. 227–243. Kang, B.T., 1997. Alley cropping—soil productivity and nutrient recycling. For. Ecol. Manage. 91, 75–81. Kass, D.L.C., Araya, J.S., Sanchez, J.O., Soto-Pinto, L., Ferreia, P., 1992. Ten years experience with alley farming in Central America. In: Paper Presented at the International Alley Farming Conference. IITA, Ibadan, Nigeria, pp. 393–402. Kass, D.L.C., Somarriba, E., 1999. Traditional fallows in Latin America. Agrofor. Syst. 47, 13–36. Kaur, B., Gupta, S.R., Singh, G., 2000. Soil carbon, microbial activity and nitrogen availability in agroforestry systems on moderately alkaline soils in northern India. Appl. Soil Ecol. 15, 283–294. King, K.F.S., 1987. The history of agroforestry. In: Steppler, H.A, Nair, P.K.R. (Eds.), Agroforestry: A Decade of Development. International Council for Research in Agroforestry (ICRAF), Nairobi, Kenya, pp. 3–13. Kwabiah, A.B., Voroney, R.P., Palm, C.A., Stoskopf, N.C., 1999. Inorganic fertilizer enrichment of soil: effect on decomposition of plant litter under subhumid tropical conditions. Biol. Fert. Soils 30, 224–231. Kwabiah, A.B., Stoskopf, N.C., Voroney, R.P., Palm, C.A., 2001. Nitrogen and phosphorus release from decomposing leaves under sub-humid tropical conditions. Biotropica 33, 229–241.

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377 Lal, R., Kimble, J.M., 2000. Tropical ecosystems and the global carbon cycle. In: Lal, R., Kimble, J.M., Stewart, B.A. (Eds.), Global Climate Change and Tropical Ecosystems. CRC–Lewis Publishers, Boca Raton, FL, pp. 3–32. Lehmann, J., Schroth, G., Zech, W., 1995. Decomposition and nutrient release from leaves, twigs, and roots of three alley-cropped legumes in central Togo. Agrofor. Syst. 29, 21– 36. Livesley, S.J., Gregory, P.J., Buresch, R.J., 2000. Competition in tree row agroforestry systems. 1. Distribution and dynamics of fine root length and biomass. Plant Soil 227, 149–161. Lose, S.J., Higer, T.H., Leihner, D.E., Kroschel, J., 2003. Cassava, maize and tree root development as affected by various agroforestry and cropping systems in Benin, West Africa. Agric. Ecosyst. Environ. 100, 137–151. Lundgren, B., 1982. Introduction [Editorial]. Agrofor. Syst. 1, 3–6. Mahboubi, P., Gordon, A.M., Stoskopf, N., Voroney, R.P., 1997. Agroforestry in the Bolivian Altiplano: evaluation of tree species and greenhouse growth of wheat on soils treated with tree leaves. Agrofor. Syst. 37, 59–77. Mazzarino, M.J., Szott, L., Mimenez, M., 1993. Dynamics of sol total C and N, microbial biomass, and water-soluble C in tropical agroecosystems. Soil Biol. Biochem. 25, 205–214. Miller, R.W., Gardiner, D.T., 2001. Soils in Our Environment. Prentice Hall, New Jersey. Miller, A.W., Pallardy, S.G., 2001. Resource competition across the crop–tree interface in a maize–silver maple temperate alley cropping stand in Missouri. Agrofor. Syst. 53, 247–259. Ministerio de Ambiente y Energ´ıa (MINAE), 2002. Publicacion No. 30962-MINAE, Costa Rica. Mugendi, D.N., Nair, P.K.R., 1997. Predicting the decomposition patterns of tree biomass in tropical highland microregions of Kenya. Agrofor. Syst. 35, 187–201. Mugendi, D.N., Nair, P.K.R., Mugwe, J.N., O’Neill, M.K., Swift, M.J., Woomer, P.L., 1999. Alley cropping of maize with Calliandra and Leucaena in the subhumid highlands of Kenya. Part 2. Biomass decomposition, N mineralization, and N uptake by maize. Agrofor. Syst. 46, 51–64. Nabuurs, G.J., Garza-Caligaris, J.F., Kanninen, M., Karjalainen, T., Lapvetelainen, T., Liski, J., Masera, O., Mohren, G.M.J., Pussinen, A., Schelhaas, M.J., 2001. CO2 Fix v. 2.0: A Manual of a Model for Quantifying Carbon Sequestration in Forest Ecosystems and Wood Products. Alterra Report XX, Wageningen, p. 48. Nair, P.K.R., 1993. An Introduction to Agroforestry. Kluwer Academic Publishers, Dordrecht, NL. Nair, P.K.R., 1998. Directions in tropical agroforestry research: past, present, and future. Agrofor. Syst. 38, 223–245. Nair, R.P.K., Nair, V.D., 2002. Carbon sequestration in agroforestry systems. In: Proceedings of the 17th World Congress on Soil Science, 14–21st August 2002. http://www.grida.no/climate/ ipcc/land use/163.htm Norby, R.J., Wullschleger, S.D., Gunderson, C.A., Hohnson, D.W., Ceulemans, R., 1999. Tree responses to rising CO2 in field experiments: implications for the future forest. Plant Cell Environ. 22, 683–714.

375

Ntayombya, P., Gordon, A.M., 1995. Effects of black locust on productivity and nitrogen nutrition of intercropped barley. Agfor. Syst. 29, 239–245. Nygren, P., 1995. Carbon and nitrogen dynamics in Erythirna poeppigiana (Leguminosae: phasseoleae) trees managed by periodic prunings. Ph.D. Thesis, Department of Forest Ecology, University of Helsinski, Finland, p. 154. Oelbermann, M., 2002. Linking carbon inputs to sustainable agriculture in Canadian and Costa Rican agroforestry systems. Ph.D. Thesis. Department of Land Resource Science, University of Guelph, p. 208. Oelbermann, M., Voroney, R.P., Kass, D.C.L., Schlönvoigt, A.M., 2003a. Aboveground carbon stocks in 10 and 19-year old tropical agroforestry systems. Agricult. Ecosys. Environ., in press. Oelbermann, M., Voroney, R.P., Schlönvoigt, A.M., 2003b. Cuantificación del carbono radicular de Erythrina poeppigiana de cuatro y diez años establecidos en callejones en Costa Rica. Agrofor. en las Amer. 11, in press. Oelbermann, M., Voroney, R.P., Schlönvoigt, A.M., Kass, D.C.L., 2003c. Decomposition of Erythrina poeppigiana leaves in 3-, 9-, and 18-year old alleycropping systems in Costa Rica. Agrofor. Syst. 62, in press. Oelbermann, M., Voroney, R.P., Kass, D.C.L., 2003d. Gliricidia sepium carbon inputs and soil carbon pools in a Costa Rican alley cropping system. Int. J. Ag. Sust. 2, 11–21. Parton, W.J., Schimel, D.S., Cole, C.V., Ojima, D.S., 1987. Analysis of factors controlling soil organic matter levels in Great Plains grasslands. Soil Sci. Soc. Am. J. 51, 1173–1179. Paul, E.A., Follett, R.F., Leavitt, S.W., Halvorson, A.D., Peterson, G.A., Lyon, D.J., 1997. Radiocarbon dating for determination of soil organic matter pool sizes and dynamics. Soil. Sci. Soc. Am. J. 61, 1058–1067. Paustian, K., Six, J., Elliott, E.T., Hunt, H.W., 2000. Management options for reducing CO2 emissions from agricultural soils. Biogeochemistry 48, 147–163. Quinlan, M.M., 1984. Mulches from two tropical tree species Erythrina poeppigiana (Walp.) O.F. Cook and Gmelina arborea Rcx. as nitrogen sources in the production of maize (Zea mays L.). M.Sc. Thesis. CATIE, Turrialba, Costa Rica, p. 74. Rao, M.R., Ong, C.K., Rathak, P., Sharma, M.M., 1991. Productivity in annual cropping and agroforestry system on a shallow Alfisol in semi-arid India. Agrofor. Syst. 15, 51–63. Rosenzweig, C., Hillel, D., 2000. Soils and global climate change: challenges and opportunities. Soil Sci. 16, 47–56. Rowe, E.C., van Noordwijk, M., Suprayogo, D., Kurniatin, H., Giller, K.E., Cadish, G., 2001. Root distributions partially explain 15 N uptake patterns in Gliricidia and Peltorphorum hedgerow intercropping systems. Agric. Ecosyst. Environ. 61, 213–224. Sánchez, J.F., 1989. Analisis de la Estabilidad y Dinamica de Sistemas de Producción de Cultivos en Callejones. M.Sc. Thesis. CATIE, Turrialba, Costa Rica, p. 161. Sánchez, P.A., 2000. Linking climate change research with food security and poverty reduction in the tropics. Agric. Ecosyst. Environ. 82, 371–383. Saggar, S., Tate, K.R., Feltham, C.W., Childs, C.W., Parshotam, A., 1994. Carbon turnover in a range of allophonic sols amended

376

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377

with 14 C-labelled glucose. Soil Biol. Biochem. 26, 1263– 1271. Saxe, H., Ellsworth, D.S., Heath, J., 1998. Tree and forest functioning in an enriched CO2 atmosphere. New Phytologist 139, 395–593. Schimel, D.S., 1994. Climatic, edaphic, and botic controls over storage and turnover of carbon in soils. Glob. Biogeochem. Cycl. 8, 279–294. Schlesinger, W.H., 1997. Biogeochemistry: An analysis of Global Change. Academic Press, London, p. 588. Schroeder, P., 1994. Carbon storage benefits of agroforestry systems. Agrofor. Syst. 27, 89–97. Schroth, G., 1995. Tree root characteristics as criteria for species selection and systems design in agroforestry. Agrofor. Syst. 30, 125–143. Schroth, G., Lehmann, J., 1995. Contrasting effects of roots and mulch from three agroforestry tree species on yield of alley cropped maize. Agric. Ecosyst. Environ. 54, 89–101. Schroth, G., Kolbe, D., Pity, B., Zech, W., 1996. Root system characteristics with agroforestry relevance of nine leguminous tree species and a spontaneous fallow in a semi-deciduous rainforest area of West Africa. For. Ecol. Manage. 84, 199– 208. Schroth, G., 1999. A review of belowground interactions in agroforestry: Focusing on mechanisms and management options. Agrofor. Syst. 42, 5–24. Schroth, G., D’Angelo, S.A., Teiseira, W.G., Haag, D., Lieberei, R., 2002. Conversion of secondary forest into agroforestry monoculture plantations in Amazonia: consequences for biomass, litter and soil carbon stocks after 7 years. For. Ecol. Manage. 163, 131–150. Seiter, S., William, R.D., Hibbs, D.E., 1999. Crop yield and tree-leaf production in three planting patterns of temperatezone alley cropping in Oregon, USA. Agrofor. Syst. 46, 273– 289. Smith, R., 1929. Tree Crops: A Permanent Agriculture. Island Press, Washington. Solomon, D., Fritzsche, F., Tekalign, M., Lehmann, J., Zech, W., 2002. Soil organic matter composition in the subhumid Ethiopian Highlands as influenced by deforestation and agricultural management. Soil Sci. Soc. Am. J. 66, 68–82. Szott, L.T., 1987. Improving the productivity of shifting cultivation in the Amazon Basin of Peru. Ph.D. Thesis, North Carolina State University, Raleigh, NC, p. 101. Thevathasan, N.V., Gordon, A.M., 1995. Moisture and fertility interactions in a potted poplar-barley intercropping. Agrofor. Syst. 29, 275–283. Thevathasan, N.V., Gordon, A.M., 1997. Poplar leaf biomass distribution and nitrogen dynamics in a poplar-barley intercropped system in southern Ontario, Canada. Agrofor. Syst. 37, 79–90. Thevathasan, N.V., Gordon, A.M., Price, G.W., 2000. Biophysical and ecological interactions in a temperate tree-based intercropping system in southern Ontario, Canada. In: Proceedings of the Fourth Biennial Meeting, North American Chapter. International Farming Systems Association, October 20–24, 1999, pp. 49–60.

Tian, G., Kang, B.T., Brussaard, L., 1992. Biological effects of plant residues with contrasting chemical compositions under humid tropical conditions: decomposition and nutrient release. Soil Biol. Biochem. 24, 1051–1060. Torn, M.S., Trumbore, S.E., Chadwick, O.A., Vitousek, P.M., Hendricks, D.M., 1997. Mineral control of soil organic carbon storage and turnover. Nature 389, 170–173. Tornquist, C.G., Hons, F.M., Feagley, S.E., Haggar, J., 1999. Agroforestry system effects on soil characteristics of the Sarapiqu´ı region of Costa Rica. Agric. Ecosyst. Environ. 73, 19–28. Torquebiau, E.F., Kwesiga, F., 1996. Root development in a Sesbania sesban fallow–maize system in eastern Zambia. Agrofor. Syst. 34, 139–211. Tubiello, F.N., Rosenzweig, C., Kimball, B.A., Pinter, P.J., Wall, G.W., Hunsaker, D.J., La Morte, R.L., Garcia, R.L., 1999. Testing CERES-wheat with free-air carbon dioxide enrichment (FACE) experiment data: CO2 and water interactions. Agron. J. 91, 247–255. Vanlauwe, B., Sanginga, N., Merckx, R., 1997. Decomposition of for Leucaena and Senna prunings in alley cropping systems under sub-humid tropical conditions: the process and its modifiers. Soil Bio. Biochem. 29, 131–137. Vanlauwe, B., Aihou, K., Aman, S., Tossah, B.K., Dields, J., Sanginga, N., Merckx, R., 2001. Leaf quality of selected hedgerow species at two canopy ages in the derived savanna zone of West Africa. Agrofor. Syst. 53, 21–30. van Noordwijk, M., Lawson, M., Soumare, G., Groot, J.H.R., Hairiah, K., 1996. Root distribution of trees and crops: competition and/or complementarity. In: Ong, C.K., Huxley, P.A. (Eds.), Tree–Crop Interactions: A Physiological Approach. CAB International, Wallingford, UK, pp. 401–421. Wang, Y.P., Rey, A., Jarvis, P.G., 1998. Carbon balance of young birch trees grown in ambient and elevated atmospheric CO2 concentrations. Glob. Change Biol. 4, 797–807. Watson, R.T., Noble, I.R., Bolin, B., Ravindranath, N.H., Verardo, D.J., Dokken, D.J., 2000. IPCC special report on land use, land-use change and forestry. http://www.grida.no/climate/ipcc/land use/ Wilken, G.C., 1977. Integrating forest and small-scale farm systems in Middle America. Agroecosystems 3, 291–302. William, C.P., Colombo, S.J., Cherry, M.L., Flannigan, M.D., Greifenhagen, S., McAlpine, R.S., Papadopol, C., Scarr, T., 2000. Third millennium forestry: what climate change might mean to forests and forest management in Ontario. For. Chron. 76, 445–462. Williams, P.A., Gordon, A.M., Garrett, H.E., Buck, L., 1997. Agroforestry in North America and its role in farming systems. In: Gordon, A.M., Newman, S.M. (Eds.), Temperate Agroforestry Systems. CAB International, Wallingford, UK, pp. 9–48. Winjum, J.K., Dixon, R.K., Schroeder, P.E., 1993. Estimating the global potential of forest and agroforestry management practices to sequester carbon. Water Air Soil Pollut. 64, 227–313. Wojtokowski, P., 1998. The Theory and Practice of Agroforestry Design. Science Publishers, Inc., Enfield, NH, 282 pp.

M. Oelbermann et al. / Agriculture, Ecosystems and Environment 104 (2004) 359–377 Wright, J.A., DiNicola, A., Gaitan, D., 2000. Latin American forest plantations: opportunities for carbon sequestration, economic development and financial returns. Am. For. 98, 20–23. Yamoah, C.F., Agboola, A.A., Wilson, G.W., 1986. Nutrient contribution and maize performance in alley-cropping systems. Agrofor. Syst. 7, 247–254.

377

Young, A., 1997. Agroforestry for Soil Management. CAB International, Wallingford, UK. Zhang, P., 1999. The impact of nutrient inputs from stemflow, throughfall, and litterfall in a tree-based temperate intercropping system, southern Ontario, Canada. M.Sc. Thesis. Department of Environmental Biology, University of Guelph, p. 130.