Background

An agroforestry system

Growing of trees in combination with other field agricultural activities, such as cultivation of crops and rearing of animals, can typically be termed as an agroforestry system. Agroforestry practices on agricultural land make an important contribution to climate change mitigation, but are not systematically accounted for in either global carbon budgets or national carbon accounting. Agroforestry has traditionally been important elements of temperate regions around the world. This practice results in a number of benefits including ensured food security, enhanced biodiversity, enrichment of an ecosystem with increased resources, and attainment of various environmental targets, e.g., maintaining atmospheric CO2 to certain limits (Ajayi et al. 2011). In addition, the trees just planted on 3–5% of agricultural lands increase farm productivity, reduce vulnerability to climate change, and decrease greenhouse gases emission (Possu et al. 2016); hence, the practice has been regarded as climate-smart agriculture (FAO 2010). Cumulatively, these benefits provide mitigation strategies to global climate change impacts (Schoeneberger et al. 2012; Cubbage et al. 2013). The main goals of agroforestry system are increasing the overall productivity and efficiency of a land use system (Nair 2005). Agroforestry systems have higher capability to store carbon in above- and belowground as compared to treeless systems (Montagnini and Nair 2004). Therefore, this system provides a sink for of atmospheric carbon. The trees, especially those with a deep rooting systems, store a large amount of atmospheric carbon in their biomass on long-term basis. Furthermore, Steinbeiss et al. (2008) reported that specific functional traits of trees with grassland species increase carbon uptake into the underground environment through resource partitioning.

Afforestation with crop production can be a strategy to control carbon fluxes in atmosphere and to mitigate climate change impacts on ecosystem (Lal 2004a; Fialho and Zinn 2014; De Moraes Sá et al. 2015; Ono et al. 2015; Muñoz-Rojas et al. 2015). Agroforestry has gained high attention in most of the developing countries for its potential for mitigating the climate variability and atmospheric CO2 sequestration (Anderson and Zerriffi 2012). This is because the climate change adaptation and mitigation objectives are highly dependent on agroforestry (Matocha et al. 2012; Stavi and Lal 2013). Therefore, agroforestry can instantaneously help addressing climate and development goals by creating “co-benefits” such as providing alternate energy source and maintaining the impact land use change on flora and fauna of a region (Watson et al. 2000; May et al. 2005; Pandey 2007; Roshetko et al. 2007; Nair et al. 2009a).

The land use and land cover changes

Land use is exercising various agricultural and non-agricultural (development) practices, whereas the land cover change increases or decreases of a given type of land use or land cover. Under this context, soil formation due to changes in vegetation land cover induced by global climate variations are at the forefront of environmental discussions (Brevik et al. 2015; Keesstra et al. 2016; Fahad and Bano 2012; Fahad et al., 2013, 2014a, b, 2015a, b, 2016a, b, c, d). The phenomena of geochemical and biological cycles and their impact on the resources, goods, and services the soils and the vegetation offer to the societies is important to understand to figure out the role of a soil system and the carbon cycle (Keesstra et al. 2012; Mol and Keesstra 2012; Decock et al. 2015; Smith et al. 2015; Berendse et al. 2015).

The carbon cycle (Fig. 1) is a key part of the environmental systems of the soil and the vegetation and their management that determine the potential use of soil for the land cover change dynamics of earth system (Gümüs and Şeker 2015; Garcia-Diaz et al. 2016; Mukhopadhyay et al. 2016). The management of the crop production is a key factor on agriculture and forest lands (Wasak and Drewnik 2015; Musinguzi et al. 2015; Turgut 2015; Novara et al. 2015) that can determine the carbon cycle and consequently changes related to it under the scenarios of climate change (Abbasi et al. 2015; Parras-Alcántara et al. 2015; Peng et al. 2015). The recent literature reported on the impact of management of the soil organic matter and the quantification of atmospheric carbon sequestration (Bruun et al. 2015; De Oliveira et al. 2015; Behera and Shukla 2015 and references therein).

Fig. 1
figure 1

The global carbon cycle with different reservoirs and the exchange of carbon between the reservoirs. The black arrows show the natural processes of carbon transfer, while the red arrows show changes driven by anthropogenic activities. The values are in units of gigatons of carbon per year (Bralower and Bice 2016)

Full size image

Greenhouse gases in atmosphere

Gaseous formation of earth’s atmosphere composes major greenhouse gases (GHGs) including water vapor, carbon dioxide (CO2), methane (CH4), nitrous oxide (NO), and ozone (O3) in addition to traces of other minor GHGs (Hussain et al. 2014). In the earth’s atmosphere, CO2 is among major GHGs. Atmospheric presence of GHGs governs average ambient air temperature, i.e., −18 °C in absence of GHGs and 15 °C in presence of GHGs (Ming et al. 2014). The earth’s climate is changing in direct response to anthropogenic GHGs emission as manifested by increase in the global average temperatures, rise of the sea levels, and melting of snow glaciers (Intergovernmental Panel on Climate Change (IPCC) 2007a; Achard et al. 2014; Liu et al. 2014; Anaya-Romero et al. 2015). The buildup of GHGs including CO2 in the atmosphere is the major cause of global climate change.

The global food production is estimated to contribute at the minimum of one third of all global anthropogenic GHGs emissions, more than twice than that of the transport sector (IPCC 2007b; Scialabba and Muller-Lindenlauf 2010). Agriculture alone contributes between 10 and 25% of annual GHGs, both directly and indirectly, through land use changes, land management, and production practices (Scialabba and Muller-Lindenlauf 2010; Smith et al. 2007). The atmospheric concentration of CO2 and other GHGs has also been increased since the industrial revolution (Cerdà et al. 2010). Though not all of released CO2 is stored in the earth’s atmosphere, considerable quantities are sequestered by land-based sinks, i.e., nearly 27.5% of CO2 productions by anthropogenic activities are taken up and recycled (Peters et al. 2012). Recent reports from the IPCC propose that even if substantial reductions in anthropogenic carbon emissions are achieved in the near future, efforts to sequester previously emitted carbon will be necessary to ensure safe levels of atmospheric carbon and to mitigate impact of climate change (Smith et al. 2014). Carbon dioxide production in the air is believed to be enhanced by anthropogenic activities especially due to deforestation and burning of fossil fuels. Increased atmospheric CO2 is considered to be the predominant reason of global climatic variability (IPCC 2007b). Furthermore, atmospheric CO2 have touched to 400 ppm (Le Quéré et al. 2015) and is predicted to reach between 700 to 900 ppm (Watson-Lazowski et al. 2016). Limiting of atmospheric CO2 concentration is possible through forest restoration and agroforestry (Montagnini and Nair 2004). There is a growing interest to identify the role of various land use systems contribution in stabilizing the atmospheric CO2 and decreasing its emissions (Murthy et al. 2013). It is anticipated that agricultural practices could modulate the increasing CO2 levels by carbon sequestration. Similarly, substitute agricultural practices not only sequester carbon but also can substitute fossil fuel consumption with biomass production. Agroforestry systems store carbon in biomass and sequester CO2 by photosynthetic processes (David and Crane 2002; Benites et al. 1999). Additionally, several studies have demonstrated that degraded land quality could be restored by adopting agroforestry system (Shazana et al. 2013; Novara et al. 2013; Tesfaye et al. 2014). Therefore, proper management of agroforestry land use systems can act as a vital option in decreasing atmospheric CO2 (Post and Kwon 2000). This system have substantially changed the land use from lone crop cultivation to adding trees and sequestering carbon in above- and belowground biomass and have fascinated the environmentalists of both developed and developing countries (IPCC 2000a; Makundi and Sathaye 2004; Takimoto et al. 2008; Gutierrez et al. 2009; Nair 2012; Poeplau and Don 2013). Therefore, agroforestry offers great potential for sequestering carbon and producing biomass for biofuels like many other land use systems (Jose and Bardhan 2012). As these have ability to capture a significant amount of atmospheric CO2 and accumulate the carbon in soil and plants. Although agroforestry systems are not primarily designed as a solution to decrease atmospheric CO2 yet it can play a major role in capturing or storing carbon in above and belowground biomass (Sathaye et al. 2001).

An agroforestry approach

Field- and home-based approaches

Agroforestry is defined as a land use system in which trees deliver biomass and environmental services. In these arrangements, various cropping systems are merged with tree plantation in the same locality for a positive change in environment and net economic returns for the farmers (Otegbeye 2002). Alao and Shuaibu (2013) defined agroforestry system as a unique arrangement of trees, crops, and animals in space and time. Agroforestry is recognized as an afforestation activity for GHGs mitigation under the Kyoto Protocol (Nair et al. 2009b). Cultivation of crops along with tree plantation in agricultural fields recovers soil fertility, prevents soil erosion, regulates water infiltration, reduces pressure on forests for fuel, and produces forage for animals (Makundi and Sathaye 2004; Yu and Jia 2014). These land use systems also maintain several other ecosystem services such as increasing, species diversity, carbon sequestration, improving soil and ecosystem health and reducing emissions of CO2 (Nair et al. 2010; Garrity et al. 2010; Hu et al. 2015; Thakur et al. 2015).

Agroforestry reduces the water losses by drainage and evaporation from soil surface and improves water use efficiency (Bayala and Wallace 2015). Rockström (1997) estimated that approximately 40% of the rainfall water in water harvesting systems was lost as evaporation while drainage caused 33–40% water losses, with only 6–16% being used by crop. Therefore, planting trees can reduce high percentages of available water lost as evaporation and drainage (Ong et al. 2006). The additional benefit of planting trees with crops is that trees do not compete with crops for the water recourses as the trees mainly absorb water beneath the root zone of crop plants below the surface soil. In arid regions, the field crop roots benefit from the soil moisture present in the rhizosphere due to hydraulic lift of water by trees (Burgess et al. 1998; Jackson et al. 2000; Hultine et al. 2003; Hao et al. 2009).

Agroforestry and soil organic carbon stocks

Several farmers in developing countries practice agroforestry and economically benefit from it (Sarkhot et al. 2007). Although several researchers reported that agroforestry land use systems have a higher capability to sequester CO2 than croplands, but it greatly depends on the environment of the area, biological, physical, and socioeconomic features of the land use system (Sanchez 2000; Sharrow and Ismail 2004; Nair and Nair 2014). Hombegowda et al. (2016) studied the effect of four land use systems: a natural forest, agriculture, and two agroforestry types of two ages (30–60 and >60 years) on carbon stocks in soils. The conversion of forest land use to agricultural system resulted in huge losses (50–61%) of original soil organic carbon (SOC) stocks in the top soil (Straaten et al. 2015). In addition to land use system, soil type can also change the SOC losses when forest land use is changed to agriculture land use (McDonagh et al. 2001; Birch-Thomsen et al. 2007). A detailed study conducted by Muñoz-Rojas et al. (2015) they evaluated the transformation of land use and land cover changes between 1956 and 2007 in Andalusia, containing the data of 1357 soil profiles. Land use changes resulted in SOC losses, specifically in Cambisols, Luvisols, and Vertisols, with the total loss of 16.8 Tg (approximately 0.33 Tg year−1). The area where forest plantation was done, increased SOC in the topsoil and it contributed 862 Mg ha−1 of SOC stocks (25%) (Muñoz-Rojas et al. 2015). Cultivation of the Vertisol for 20 years resulted in 40% lower SOC content in comparison to area under forest land use and during this time about 95% of the forest originated SOC was lost in area under cultivation. In contrast, Ultisol cultivation resulted only 20% lower SOC than soil under forest land use and only 30% of the forest originated SOC was lost (Bruun et al. 2015). However, Ferreira et al. (2016) demonstrated that land use change from forest to cropland caused carbon losses from the soil but conversion of forest to pastures land might increase the net carbon in the soil due to high carbon and nitrogen stocks, higher soil microbial biomass and lower respiratory quotient results in net carbon sequestration in the soil.

The establishment of agroforestry as home garden and coffee production on agriculture land caused SOC stocks to rebound to near forest levels. On the other hands, planting mango and coconut trees increased SOC stocks slightly above the agriculture SOC stocks. The authors have found a strong correlation between tree species diversity in home garden and coffee agroforestry and SOC stocks (Cadotte 2013). Therefore, judicious use of agroforestry practices should be made to enhance the system use efficiency for agricultural productivity on sustainable basis in combination with meeting other societal needs from forestry (Fagbemi 2002). This is a win–win situation both in terms of meeting human demand as well as environment sustainability for longer period of time (Alao and Shuaibu 2013).

Atmospheric carbon sequestration

Carbon is present in various forms in different parts of the Earth. In the atmosphere carbon present as CO2 converted into various organic compounds through photosynthesis. The photosynthetically metabolized CO2 is converted back in to CO2 through respiration. Some of the carbon from the atmosphere is absorbed by the ocean that subsequently is converted into sedimentary rocks, and much later, this carbon may be released to the atmosphere. So carbon moves around, it flows from place to place and circulates among various components in the cycle as shown in Fig. 1. The continued buildup of atmospheric CO2 over the last century with projected rise in near future (Paustian et al. 2000) has raised serious concern among the environmentalists. The increased CO2 in the atmosphere have some benefits as it serves as a stimulant to improve plants growth and productivity (Schaffer et al. 1997; Keutgen and Chen 2001). However, climate extremes including rising temperatures and uneven distribution of rainfall are also associated with an increased concentration of atmospheric CO2 (USDA NRCS 2000; Abbas 2013; Abbas et al. 2014). A rich literature has been produced on carbon sequestration especially during the last two decades. However, significant differences exist among individuals about the role of increasing concentration of atmospheric CO2 and the related pros and cons to the ecosystem. Persistent increase of carbon storage in soil and plant material and in the sea is termed as carbon sequestration (Hutchinson et al. 2007; Pandey et al. 2016). According to United Nations Framework Convention on Climate Change, carbon sequestration is the secure storage of CO2 in soil and plant. It depends on the metabolic conversion of CO2 into long-lived, carbon containing materials (through photosynthesis), a process that is called bio-sequestration (DOE/SC-108, US Department of Energy 2008). Carbon sequestration is successful when carbon storage resulting from land management and/or conservation practices exceeds carbon losses (IPCC 2007a; Smith et al. 2014). Carbon sequestration is possible through a range of processes, occurring naturally in plants and soils. Recently, carbon sequestration and decreased emissions from circumvented deforestation have received more attention as a method to reduce the buildup of GHGs in the earth atmosphere (Sedjo and Brent 2012). Carbon sequestration happens in two main segments of agroforestry systems: belowground and aboveground. The aboveground segment is described as specific plant components (such as stem and leaves of herbaceous plants and trees), while the belowground segment contains roots and soil microorganisms, and soil organic carbon present in different soil horizons. Due to net positive contribution of agroforestry to climate change system, the system has become habitual for the term carbon sequestration. The belowground biomass carbon is more stabilized in the soil due to its interactions with soil particles (Rasse et al. 2005) and its slow decomposition rate in comparison to above ground biomass (Cusack et al. 2009). As Scheu and Schauermann (1994) reported that relative contribution of belowground biomass by Fagus sylvatica L. to SOC was 1.55 times higher than that above ground biomass (Johnson et al. 2006).

Researchers have demonstrated that carbon sequestration to stabilize SOC in urban and agricultural soils is one of numerous options to reduce the atmospheric concentration of CO2 (Bruce et al. 1999; Pouyat et al. 2002; Leified 2006; Pataki et al. 2006; Pickett et al. 2008; Blanco-Canqui and Lal 2008). Additionally, SOC is due to historic buildup of humus in the soil. When soil humus reaches a point of stability, it results in long-term storage of carbon in soil (Whitehead and Tinsley 2006). If soil remains undisturbed, soil humus can retain carbon for an average lifetime of hundreds to thousands of years (Holmén 2000).

Carbon sequestration through agroforestry depends on cropping systems that define land cover change (Thevathasan and Gordon 2004; Steinbeiss et al. 2008). Agriculture-based land use changes contribute approximately 20% of the total CO2 sequestration by anthropogenic sources (Dumanski and Lal 2004). The top 30 cm of soil layer has average SOC value reaching approximately 15 Mg ha−1; however, during cultivation, about 50–75% of this carbon is released to the atmosphere within the first 20 years in the tropical regions and 20–30% in temperate regions (De Blécourt et al. 2013; Chiti et al. 2014). Nevertheless, by adopting soil conservation practices on arable soils, considerable amount of this carbon can be prevented from emission through soils. A huge carbon sequestration potential by major croplands has been estimated by Dumanski and Lal (2004) as shown in Table 1.

Table 1 Possibilities of world’s major croplands for carbon sequestration (Dumanski and Lal (2004)
Full size table

Montagnini and Nair (2004) reported that the agroforestry land use systems with higher net primary carbon assimilation treeless land use system returned a greater portion of plant biomass back to the soil and it had the greater potential to increase soil organic carbon. In addition, agroforestry systems have a higher ability to store carbon than field crops and grasslands (Kirby and Potvin 2007). Moreover, tree, shrub, and pastures residues in agroforestry systems increase SOC (Abbasi et al. 2015). Agroforestry also offers a great scope of economic development of rural people.

Mitigation of climate change impacts through agroforestry

Carbon sequestration by agroforestry practices has been considered beneficial in climate change impact mitigation. Agroforestry has diverse advantages such as the plants provide a considerable sink for atmospheric carbon due to their high growth rate and quick biomass productivity. The trees in agricultural land use systems can enhance the carbon sequestered in farm soils reserved to agriculture, while simultaneously allowing for the growing of food crops (Kursten 2000).

Soils act as a sink to store carbon from the atmosphere for longer period of time. Based on historic global estimates carbon stocks and emissions, soil provides a useful carbon sink for necessary solution to environmental problems (Lal 2004c, 2008). Since agriculture occupies over one third of arable land globally (World Bank 2015); therefore, agroforestry presents a great potential for increased sequestration of carbon in agricultural lands.

Planting trees with nitrogen fixing capability may also increase biomass production. Since the sequestered carbon and nitrogen in soils has complex interaction. Study conducted in Malawi and Zambia showed that cultivating maize with Gliricida—a nitrogen fixing trees—has 42% higher yields than non-fertilized fields and similar to fields receiving 92 kg N ha−1 (Sileshi et al. 2012). Moreover, integrating Gliricidia trees during fallow periods between crops resulted in 55% increase in sorghum productivity (Hall et al. 2005). However, strategies must focus on synchronizing legume tress with crop nitrogen demand to regulate gaseous and leaching losses of N from the soil (Rosenstock et al. 2014). A study is needed to further investigate interactions of SOC with various forms of nitrogen produced during its mineralization and immobilization under agroforestry systems (Nair et al. 2009a; Gärdenäs et al. 2011). Among the other reasons, positive effect of trees on SOC sequestration may become obvious due to modifications in belowground C stocks (Laganière et al. 2010). The higher SOC sequestration potential under agroforestry may be reflected by higher amount of SOC in deeper mineral soil layers in comparison to fields with only crop cultivation (Nepstad et al. 1994; Jobbágy and Jackson 2000). Furthermore, the tree species modify the microbial community structure and diversity in soil that may also enhance soil carbon sequestration. However, detailed investigations are needed to further elaborate the mechanisms associated with SOC sequestration in managed agroforestry systems. Because some of the planted tree species may have negative impacts on crops due competition for water (Burgess et al. 1998), more negative effects may arise due to allelochemicals (Jose et al. 2004; Inderjit 2002). Most of the tropical agroforestry have negative allelopathic effects on food and fodder crops and vice versa may also occur. Therefore, species mixtures with no or positive allelopathic effects on the companion crops must be created in agroforestry systems (Rizvi et al. 1999).

Estimation of carbon sequestration in agroforestry system

Some efforts have been carried out to evaluate the global potential of agroforestry systems as a sink for carbon. Approximately, a carbon sequestration potential of 391,000 MgC year−1 by 2010 and 586,000 MgC year−1 by 2040 by converting 630 million ha of unproductive croplands into agroforestry land use system have been estimated for 50-year period with its calculated range between 1.13 and 2.24 PgC year−1 globally (Jose 2009; Dixon 1995). A comprehensive study has been conducted by Zomer et al. (2016) in which they have estimated the contribution of agroforestry in carbon sequestration at global, regional and at country level negated by IPCC for estimating carbon biomass in agricultural systems (Table 2). According to IPCC estimates total biomass carbon through tree distribution on agricultural land broadly followed bioclimatic zones (Fig. 2) and the high tree cover (>45%) was found in the humid regions (Zomer et al. 2016). Overall, the amount of area classed as agricultural is the globally the carbon stored in above- and belowground biomass is 11.1 PgC in agricultural lands, when agricultural area is ~22.2 million km2 (Bartholomé and Belward 2005). However, when tree in agro systems are considered in carbon storage the agricultural land has four times higher values (45.30 PgC) than the default values estimated by IPCC (Zomer et al. 2016). In addition, the authors has also found that there was 2% an additional increase tree cover between 2000 and 2010, resulting in an increase of >2 PgC (or 4.6%) biomass carbon (Fig. 3).

Table 2 Total biomass carbon on agricultural land (in PgC, and as a percentage of the total biomass carbon from 2000 and 2010) globally, and the contribution by trees to biomass carbon on agricultural land
Full size table
Fig. 2
figure 2

Global map of average biomass carbon per hectare on agricultural land in 2000 and 2010 (t C per ha). Source Zomer et al. 2016

Full size image
Fig. 3
figure 3

Global map in average biomass carbon from 2000 and 2010 per hectare on agricultural land (t C per ha). Source Zomer et al. 2016

Full size image

The carbon sequestration capacity differs across the geography of the area and plant species used in agroforestry system (Newaj and Dhyani 2008). However, evidence shows that agroforestry land use can act as both source and sink of carbon in the environment (Montagnini and Nair 2004; Ajayi et al. 2011). For example, agrisilvicultural systems in which crops and trees are cultivated together act as net sinks of CO2 while agro silvipastoral systems are possible net sources of greenhouse gases (Montagnini and Nair 2004). In contrast, Mangalassery et al. (2014) reported that silvipastoral system sequestered 36–60% higher CO2 compared to the tree system and 27–71% more in comparison to the grasslands. Silvipastoral system involving trees and grasses sequestered more soil organic carbon compared with only trees or pasture containing systems. Carbon sequestering potential of different agroforestry systems varies depending on species composition, soil and climate. Similarly, tropical regions have higher vegetation carbon sequestration potential than temperate agroforestry regions. The sequestered carbon in the above- and belowground biomass is highly variable of an agroforestry system and is usually much higher than treeless land use system (Nair et al. 2009a; Fialho and Zinn 2014). Potential for sequestering carbon in aboveground components of agroforestry systems is estimated to be 2.1 × 109 MgC year−1 in tropical and 1.9 × 109 MgC year−1 in temperate biomes (Oelbermann et al. 2004). The IPCC report suggest that even after achieving global targets of carbon sequestration, efforts to sequester previously emitted carbon will remain necessary to achieve safe levels of atmospheric concentration of carbon for mitigating climate change impacts (Smith et al. 2014).

Higher soil organic carbon (SOC) in agroforestry land use systems can be particularly obtained by enhancing the amount of carbon returned to the soil and by strengthening soil organic matter (Lal 2005; Sollins et al. 2007). Agroforestry land use systems can also be managed by increasing SOC reservoir in the soil through avoiding burning and minimizing soil disturbance by minimum or zero tillage practices and by erosion control (Soto-Pinto et al. 2010). As Sá et al. (2015) compared the tillage system in relation to SOC losses and concluded that no-till systems have a large potential to decrease soil degradation and SOC decline in comparison to conventional tillage systems.

Soil organic carbon pools do not only reduce the net CO2 in the atmosphere but also play an important role in maintaining soil productivity by improving nitrogen (N) cycling in soil–plant systems (Yu and Jia 2014; Abbasi et al. 2015). For every Mg increase in profile SOC stock, an increase yield of 0.17 (pearl millet), 0.14 (cluster bean), and 0.15 (castor) Mg ha−1 year−1 were observed (Srinivasarao et al. 2014). Soil organic carbon is reliable and field-based soil quality indicator for assessing yield (Carter et al. 2003; Lal 2006). In addition, researchers reported that trading sequestered carbon was a viable economic opportunity for practitioners of agroforestry for the subsistence farmers in low-income countries (Nair et al. 2010). While agroforestry is documented as having the greatest capability for carbon sequestration, IPCC (2000b) examined land uses as described in Table 3.

Table 3 Potential of carbon sequestration by 2040
Full size table

Terrestrial biosphere plays an important role in global carbon cycle; the environmental changes are continuously changing global terrestrial carbon uptake. Carbon is continually being cycled between different pools such as soil, atmosphere, and oceans. In fact, the total amount of carbon remains constant while increased amount of carbon into a pool is balanced by an addition of equal amount of carbon into another pool. Carbon budget is actually a list of all transformations and changes occurring in various pools in which carbon is stored. Presently, a budget of the earth’s carbon cycle shows an imbalance among of various carbon pools that is mainly caused by burning of fossil fuel and change in land use system. Resultantly, CO2 is building up in atmosphere. According to the global carbon budget report of 2015 (Fig. 4), during the year 1870 to 2014, burning of fossil fuel and land use change added 1465 and 549.6 Gt of CO2 in atmosphere, respectively, while 545 ± 55 Gt of this added CO2 is recycled by atmosphere (230 ± 5 Gt), ocean (155 ± 20 Gt), and the land (160 ± 60 Gt) (Le Quéré et al. 2015).

Fig. 4
figure 4

Global average carbon budget (Gt CO2 per year) for the decade 2005–2014 (Le Quéré et al. 2015) design by GBP

Full size image

Globally, the soils store 2500 billion tons of carbon. It is more than that is stored in atmosphere (780 billion tons) and plants (560 billion tons). Approximately 5000–10,000 billion tons of carbon is stored by fossil fuels originated from fossilized plants and animals store (Le Quéré et al. 2015). Plant functioning in terms of photosynthesis stabilizes atmospheric CO2 and releases the oxygen to the atmosphere. Almost 40% of the photosynthetically stabilized CO2 is released by plant in the form of roots exudates that provide food to soil microbes. The soil microbes depend on these root exudates and convert simpler organic compounds into complex, stable forms of soil carbon, such as humus (Ahmad et al. 2009: Le Quéré et al. 2015).

Carbon sequestration in above- and belowground biomass

The concept of the carbon sequestration contains ambiguity particularly with the concept of “long-lived” pools. In agroforestry systems, carbon stocks are represented as synonym to carbon sequestration. The carbon sequestration determinations are simple mathematical calculations, in which aboveground biomass is assessed from general allometric equations while, belowground biomass is usually 30% of aboveground biomass, whereas 50% of the total plant biomass is considered as carbon stock or sequestered carbon. Complex mixtures of agricultural crops and trees are widely used for estimating aboveground carbon sequestration potential. Carbon constitutes almost 45 to 50% of stem/branches biomass and 30% of foliage dry weight (Shepherd and Montagnini 2001; Schroth et al. 2002). The carbon sequestration in soils differs extensively by depending on the agroforestry system. However, in this regard, the literature, for instance, Oelbermann et al. (2006), Amézquita et al. (2005), and Nair et al. (2009b), reported that SOC pools range from 1.3 MgC ha−1 in the top 40 cm to 173 MgC ha−1 in the top 100 cm of soil layer with 13-year-old alley cropping practices in southern Canada and 10- to 16-year-old silvopastoral systems at the Atlantic Coast of Costa Rica, respectively. Soil carbon stocks in croplands and forests under slash-and-burn systems showed that intensive cropping with short-term fallow systems in sub-humid tropics have relatively lesser carbon sequestration potential than slash-and-burn systems of the humid region of Brazil (Mutuo et al. 2005). Additionally, physical and biotic factors, as well as on management practices determine the carbon sequestration capacity.

Modeling the carbon sequestration

Model is a representation of system that allows investigation of properties of the system and prediction of future outcomes of real systems. Models are used in a variety of scientific disciplines ranging from physics and chemistry to ecology and the Earth sciences. Models are also used in food production systems support the farmers in planning day-to-day crop management practices on farms, guiding the ways to alleviate rural poverty, and predicting the effects of climate variability on food security issues (Thornton et al. 1997; Hochman et al. 2009; Webber et al. 2014). Models are helpful regarding strategic decisions and can simulate the productivity of farms and food system in various environmental conditions (Holzworth et al. 2014). Numerous models can be used to simulate the potential of SOC sequestration (Rickman et al. 2001; Verburg et al. 2002; Smith et al. 1993, 2008; Verburg and Overmars 2009; Debolini et al. 2015). Changes in the SOC pool can be measured on small scale and on large regional scales. For a small (plot) scale (Bruce et al. 1999), the direct measurement is an efficient technique (Qian et al. 2003). For large (regional) scales measurements, mathematical models of SOC have been established and extensively used to study SOC dynamics worldwide (Post et al. 2004; Qian et al. 2003; Smith et al. 2008). Models have been used to evaluate the effect of management practices on the changes in the SOC pool (Blanco-Canqui and Lal 2004; Bruce et al. 1999; Lal 2004b).

The RothC (Rothamsted model) and CENTURY models are the most commonly used as tool for simulation of soil carbon (Coleman and Jenkinson 1996). The RothC model application is based on the long-term trials conducted on Rothamsted research station to study the cycling of organic matter in soil. Although the assumptions or variables of the models are simple, the models cannot appropriately and accurately predict the carbon cycling in tropical agroforestry systems. The CENTURY model is used for the cycling of carbon and its interaction with plant species and management practices such as tillage and agricultural system. Several models in agroforestry have been used including; SCUAF, HyPAR, Hi-SAFE/Yield-SAFE and WaNuLCAS; however, their use has remained limited due to inflexibility, restricted capacity to simulate various interactions and lack of model calibration in different scenarios (Luedeling et al. 2016).

To expand the applicability of this model for estimating carbon sequestration at global scale, it must also consider the agroforestry during the prediction of carbon cycling in the system. Several scientific models have been developed to forecast the response of soil organic carbon. There are some complications in obtaining information which are necessary for the models (Nair et al. 2010). These complications reduce applicability of these models to integrate agroforestry system.

Few attempts have been made to integrate agroforestry systems into existing models or models that have been developed with agroforestry in target. For example, Palma et al. (2007) modeled silvoarble agroforestry in Europe, Negash and Kanninen (2015) used the CO2FIX model to predict soil carbon sequestration, and Francaviglia et al. (2012) used the RothC model to simulate an agro-silvopastoral system. Palma et al. (2007) used nitrogen leaching, soil erosion, landscape biodiversity, and carbon sequestration as indicators that are assessed using Yield-SAFE (from “Yield Estimator for Long term Design of Silvoarable AgroForestry in Europe”) while soil erosion was simulated using the revised universal soil loss equation (Renard et al. 1997). In Ethiopia, the CO2FIX model was used to predict the effects of three agroforestry systems on organic carbon pools in soil. Model validated that long-term (10–40 years) carbon sequestration was in the range of measured biomass for two agroforestry system (Enset-tree and Enset-coffee-tree systems), but significantly differed for the tree-coffee system (Negash and Kanninen 2015). The authors concluded that the prediction of the biomass carbon stocks could be improved by having more accurate input parameters for the model. Basic problem in application of existing modeling framework and sub-models in agroforestry system is the complexity of simulating tree growth for different tree species. The existing models have deficiency to simulate developing foliage, wood, branches, and roots. Modifications in the existing models are necessary to make them compatible with crop growth models (Pinkard et al. 2010; Almeida et al. 2010; Ghezehei et al. 2015). For instance, some tree models are not able to simulate at a daily time step basis. Notable exceptions are present in tree sub-models such as in APSIM, CABALA and 3PG. Therefore, a rapid progress in reliable modeling and its calibration for tree and crop agroforestry systems are needed for evaluating and predicting future outcomes of site specific agroforestry systems potential to sequester carbon under changing climate.

Conclusions

The evidences suggest that conversion of forest to agriculture land use results in land degradation with huge losses of soil organic carbon stocks. Cultivation of land releases about 20–70% of the stored carbon within two decades depending on the climatic conditions. Resultantly, CO2 is building up in atmosphere. Agroforestry systems retain much higher quantities of carbon in above and belowground biomass in comparison to crop and grazing land use systems. At global scale, 630 million ha of unproductive croplands could be used for agroforestry as part of an ecological engineering practice to potentially sequester 586,000 MgC year−1 by 2040. Moreover, in current global and national carbon monitoring protocols, there is a need to incorporate agroforestry in carbon stocks to precisely estimate the contribution of this neglected pool. To simulate the potential of agroforestry systems in sequestering carbon new models are needed that can precisely predict net uptake of atmospheric CO2 compared to treeless systems especially under the IPCC scenarios of projected global climate change.