Keywords

Introduction

India is a physiographically diverse and geographically large country with varied ecologies. Rich natural resource endowments in terms of soil, plant, animal, and fish wealth make India and the contiguous areas of South Asia a mega-biodiverse region. The National Bureau of Soil Survey and Land Use Planning (India), based on soil, bioclimatic, and physiographic features (Sehgal et al. 1992), has divided the country into 20 agroecological regions (Fig. 1), which broadly fall under arid, semiarid, subhumid, humid-perhumid, and coastal ecosystems. Land-use systems differ profoundly across these regions but agroforestry dominates in most parts. The Indian farmers, as their counterparts elsewhere, have domesticated fruit trees and other agricultural crops over millennia, primarily to meet their subsistence requirements. The tropical homegardens , which represent a complex integration of diverse trees (Fig. 2) with understory crops performing several production and service functions, are a case in point (Kumar et al. 2012). Indeed, the biophysical heterogeneity and climatic variability of the country affect the choice of tree and crop species and their productivity, implying profound variability in the nature and composition of agroforestry practices in India (Tejwani 1994; Puri and Panwar 2007). India is also one of the early countries to launch a national initiative on agroforestry research; indeed, as early as in 1983, it started the All India Coordinated Research Project on Agroforestry (Chinnamani 1993).

Fig. 1
figure 1

Agroecological regions of India. 1. Western Himalayas (cold arid), 2. Western Plains and Kutch Peninsula (hot arid), 3. Deccan Plateau (hot arid), 4. Northern Plains (Upper Gangetic; semiarid to subhumid), 5. Northern Plains (Rajasthan Upland and Gujarat Plains; hot semiarid), 6. Northern Plains (Middle Gangetic Plains; hot semiarid to subhumid), 7. Deccan Plateau (Malwa Plateau, Gujarat Plains, and Kathiawar peninsula; hot, semiarid with moderately deep black soils and length of growing period (LGP) 120–150 days), 8. Deccan Plateau (hot semiarid with mixed red and black soils and LGP 120–180 days), 9. Deccan Plateau (hot semiarid with red loamy soils and LGP 150–210 days), 10. Eastern Plateau (Satpura Range and Mahanadi Basin; hot subhumid), 11. Eastern Plateau (Bundelkhand Upland; hot subhumid with red and yellow soils and LGP 120–180 days), 12. Eastern Plateau (hot subhumid with red and lateritic soils and LGP 150–210+ days), 13. Northern Plains (Lower Gangetic; hot, subhumid), 14. Western Himalayas (warm to hot subhumid to humid), 15. Bengal basin (hot, subhumid), 16. Assam and North Bengal Plains (warm humid to perhumid), 17. Eastern Himalayas (warm perhumid), 18. North Eastern hills (Purvanchal; warm perhumid), 19. Eastern Coastal Plains and Islands of Andaman and Nicobar (hot subhumid), and 20. Western Ghats (Coastal Plains and Western Hills; hot humid to perhumid). Reprinted/adapted by permission from the National Bureau of Soil Survey and Land Use Planning, Nagpur (source: http://www.bhoomigeoportal-nbsslup.in/)

Full size image
Fig. 2
figure 2

A Kerala homegarden with a multistrata arrangement of coconut palms (Cocos nucifera), banana (Musa spp.), and other species (photo: BM Kumar)

Full size image

Since the late twentieth century, the phenomenon of “climate change” or “global warming” has been attracting global attention at a scale unparalleled in the history of humankind. Scientists, policy makers, and the general public continue to grapple with the adverse impacts of climate change and in figuring out strategies for mitigating the same. It is very likely that climate change may cause unprecedented shifts in global weather patterns producing a range of effects from threats to food security to rising sea levels that increase the risk of catastrophic flooding. India’s average temperature has risen by around 0.7 °C during the 1901–2018 period and it is likely to increase further by approximately 4.4 °C by 2100 (relative to the 1976–2005 average; Krishnan et al. 2020). It is widely recognized that climate change is caused by rise in the atmospheric concentrations of the so-called greenhouse gases (GHGs) such as carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). The atmospheric concentration of CO2, a prominent GHG, which accounts for 76% of the total global GHG emissions, has increased at unprecedented rates from the pre-industrial concentration of about 280 ppm to the current level of approximately 410 ppm (https://www.esrl.noaa.gov/gmd/ccgg/trends/gl_trend.html). The principal anthropogenic factors contributing to the increase in atmospheric CO2 levels include the burning of fossil fuels such as coal, gas, and oil for industrial and other purposes, and agriculture, forestry, and other land uses (AFOLU), including deforestation. The average decadal growth rate of CO2, which was 2.0 ppm per year in the 2000s, had surged to 2.4 ppm per year during the 2010–2019 period (https://www.co2.earth/co2-acceleration). Significantly, India is the third largest emitter of GHGs and accounts for 7% of total GHG emissions in the world as per the 2018 emission data (https://www.ucsusa.org/resources/each-countrys-share-co2-emissions).

Carbon sequestration is a key strategy for reducing atmospheric concentrations of CO2, and thereby mitigating global warming. It is a process of storing atmospheric CO2 or other forms of carbon (C) in long-standing pools. The United Nations Framework Convention on Climate Change (UNFCCC) describes it as “the process of removing C from the atmosphere and depositing it in a reservoir, or the transfer of atmospheric CO2 to secure storage in long-lived pools” (UNFCCC 2007). Green plants—especially woody perennials—and soil play a central role in this. Dubbed as biological carbon sequestration, plants assimilate atmospheric CO2 through photosynthesis and store the products of photosynthesis in their parts. The soil also is a major C sink as organic matter can remain in the soil for extended periods. Forestry and agroforestry systems (AFS) play a major role in biological carbon sequestration and stabilization of atmospheric GHG levels. Ever since climate change became a matter of stark global concern, agroforestry has received immense importance as a land management strategy with considerable potential for reducing atmospheric CO2 levels. The average carbon sequestration potential (CSP) of agroforestry in India has been estimated to be 25 Mg C ha−1 over 96 million ha (Sathaye and Ravindranath 1998) and agroforestry figures prominently in the country’s climate change mitigation strategies (https://www4.unfccc.int/sites/ndcstaging/PublishedDocuments/IndiaFirst/INDIAINDCTOUNFCCC.pdf). There are, however, considerable variations in the CSP of agroforestry across different regions and land-use systems and based on the method of estimation. This chapter examines the range of AFS by agroecological regions of India and their potential to sequester atmospheric CO2 and thus mitigate global warming. Such information can help focus attention on promising AFS and in adopting appropriate stand management practices including choice of species for enhancing the potential of biological carbon sequestration and for evolving national climate change mitigation strategies, which are cost effective.

Agroforestry: A Cardinal Feature of the Indian Landscape

India is regarded as the cradle of agroforestry with diverse kinds of AFS (Kumar et al. 2012). These include the tropical, subtropical, and temperate AFS. India, with a geographical area of 329 million hectares, features 20 diverse agroecological regions each with an array of AFS (Table 1). Many of these are indeed traditional systems, practiced since time immemorial. For instance, homegardening and rearing of silkworm (Bombyx spp.) and lac insect (Kerria lacca) were practiced in the Indian subcontinent during the epic era of Ramayana and Mahabharat (7000 and 4000 BCE, respectively; Puri and Nair 2004). The travelogue of Ibn Battuta (Persian traveler; 1325–1354 CE) provides the earliest literary evidence of agroforestry from peninsular India and it mentions that in the densely populated and intensively cultivated landscapes of Malabar Coast, coconut (Cocos nucifera) and black pepper (Piper nigrum) were prominent around the houses (Randhawa 1980). The ecoclimatic situations under which agroforestry is practiced in India are also correspondingly diverse and range from the humid tropical valleys through to the high-elevation temperate regions and from humid tropical forests to the semiarid and arid drylands, including both irrigated and rain-fed ecosystems.

Table 1 Major agroforestry systems and practices in different agroclimatic regions of India
Full size table

The predominant Indian AFS include agrisilviculture involving poplar (Populus deltoides; Fig. 3); Eucalyptus spp.; plantation agriculture involving coffee (Coffea spp.; Fig. 4), tea (Camellia sinensis; Fig. 5), cacao (Theobroma cacao), and spices (e.g., black pepper, cardamom, or Elettaria cardamomum) in association with a wide spectrum of trees (planted as well as trees in the natural forests); betel vine (Piper betel L.) + areca palm (Areca catechu); intercropping systems with coconut, Para rubber (Hevea brasiliensis), and other trees; commercial crop production under the shade of trees in natural forests (e.g., cardamom; Fig. 6); homegarden systems; and parkland systems. Table 1 provides a detailed account on this, agroecological region-wise. Deliberate growing of trees on field bunds (risers) and in agricultural fields as scattered trees and the practice to utilize open interspaces in the newly planted orchards and forests for cultivating field crops are also widespread in the Indian subcontinent (Singh 1987). In the relatively bigger landholdings of Himachal Pradesh, agri-horticulture is widespread, and in the northern and southern aspects, apple trees (Malus domestica) dominate. Growing arable crops in association with alder (Alnus nepalensis) is a remunerative AFS in the northeastern hill region of the country. Indeed, alder-based production system is an outstanding example of sustainable land use that stood the test of time in many parts of eastern Himalayas. Kumar et al. (2018) recently reviewed the literature on agroforestry in the Indian Himalayan region.

Fig. 3
figure 3

Agroforestry systems involving poplar (Populus deltoides), turmeric (Curcuma longa), mango (Mangifera indica; pruned trees), and litchi (Litchi chinensis) in Yamunanagar district, Haryana; note the systematic arrangement of different components (photo: BM Kumar)

Full size image
Fig. 4
figure 4

Coffee (Coffea spp.) agroforestry in Wayanad, Kerala; shade-loving coffee plants are raised in the understory of areca palms (Areca catechu) (photo: BM Kumar)

Full size image
Fig. 5
figure 5

Tea (Camellia sinensis) + silver oak (Grevillea robusta) trees (for partial shade) in Idukki district, Kerala (photo BM Kumar). Reprinted/adapted by permission from Springer (South Asian Agroforestry: Traditions, Transformations, and Prospects; Kumar et al. 2012)

Full size image
Fig. 6
figure 6

Cardamom (Elettaria cardamomum) with diverse kinds of shade trees in Idukki district, Kerala; principal trees include Vernonia arborea, Artocarpus heterophyllus, Actinodaphne malabarica, and Persea macrantha (photo: BM Kumar)

Full size image

The traditional land-use systems, however, have been transformed over time—owing to the interplay of socioeconomic and technological factors. In particular, agricultural transformations brought about by market economies in the past, especially the incorporation of exotic commercial crops (e.g., Hevea brasiliensis), have led to the decimation of many traditional land-use systems (Kumar 2005). For example, the homegardens that constituted a predominant land-use activity in the subcontinent, of late, have been showing symptoms of decline in some localities (Guillerme et al. 2011)—owing to rising population pressure and policies oriented towards land-use intensification to meet the rising demands for food grains (e.g., promoting monospecific production systems).

Environmental concerns such as global warming, land degradation, erosion of biodiversity, loss of wildlife habitats, and increased nonpoint source pollution of ground- and surface water, however, have provided impetus for the development and adoption of agroforestry around the world. Of late, economic incentives to the land managers have also acted as a major driver for promoting agroforestry. The poplar-based agroforestry in northern India, especially in the lowland “Tarai” areas at the base of the Himalayas, is a case in point (Fig. 7). An estimated 317,800 ha has been planted with P. deltoides in the country, of which 60% are block plantations and 40% are boundary plantations (National Poplar Commission of India 2012–15). Woodlots of other fast-growing trees such as eucalypts (Eucalyptus spp.), leucaena (Leucaena leucocephala), casuarina (Casuarina equisetifolia), mangium (Acacia mangium), Australian wattle (Acacia auriculiformis), maharukh (Ailanthus triphysa), and Malabar neem (Melia dubia) are also becoming increasingly popular among farmers in several parts of India.

Fig. 7
figure 7

Poplar (Populus deltoides) trees (leafless during winter) and understory wheat (Triticum aestivum) in Pantnagar, Uttarakhand (photo: BM Kumar)

Full size image

Area Under Agroforestry in India

Although AFS abound in India, precise quantitative estimates on the extent of area under agroforestry are lacking—presumably because of the nonavailability of proper procedures for delineating the area influenced by trees in a mixed stand of trees and crops (Nair et al. 2009a). While in the multistrata systems (e.g., homegardens, shaded perennial systems, and intensive tree intercropping) the entire area occupied by such tree-crop combinations can be reckoned as agroforestry, most other agroforestry systems are rather extensive, where the components, especially trees, are not planted at regular spacing or density; for example, the parkland system and extensive silvopastures in central and northern India. The problem is acute in the case of practices such as windbreaks and boundary planting where the trees are planted at wide intervals or on farm boundaries. In the sequential agroforestry systems such as improved fallows and shifting cultivation, the beneficial effect of woody vegetation (in the fallow phase) on the crops in the sequence (in the cropping phase) may last for a variable length of time (years).

Given the diversity of AFS in India and the complexity of its components, it is a formidable task to determine the area under agroforestry. Nonetheless, some attempts have been made in this direction. Dhyani et al. (2013), using the databases of agricultural, horticultural, and forestlands of the country, deduced the area under agroforestry as 25.32 m ha, or 8.2% of the total geographical area of India with Maharashtra, Gujarat, and Rajasthan ranking high among the states. In another attempt, Rizvi et al. (2014), using geospatial techniques, estimated the area under agroforestry in India as 14.46 m ha and the potential area as 17.45 m ha. Forest Survey of India (FSI 2013), using digital interpretation of remote sensing data, however, estimated it as 11.54 m ha. Given the lack of consistency among the available estimates and the need to evolve climate change mitigation strategies through land-use management, it is imperative to estimate the area under agroforestry in India more precisely; however, such efforts are still rudimentary.

Agroforestry for Climate Change Mitigation and Adaptation

Agroforestry provides an excellent opportunity for combining the twin aims of climate change mitigation (technological changes and substitution that reduce GHG emissions by averting emissions and sequestering GHGs) and adaptation (evolving approaches to reduce the harmful effects of climate change). In addition to its potential for reducing atmospheric CO2 levels, AFS play an important role in reducing vulnerability of agricultural production systems to climate change (i.e., imparting increased resilience); they also increase livelihood security of the dependent populations. Given such advantages, the importance of promoting agroforestry in the country cannot be overemphasized. In particular, there is scope for conversion of wastelands and grasslands to agroforestry, which according to IPCC (2007) has huge potential to absorb CO2 from the atmosphere. There are about 120 million hectares of degraded lands in India (ICAR-NAAS 2010) and a significant chunk of that could probably be converted into agroforestry. While the potential for agroforestry in India is enormous, there are also challenges such as dearth of quality planting materials, lack of credit and marketing facilities, meager insurance cover, and weak extension, which hamper the adoption of AFS. To capitalize on the ecological and production functions of agroforestry, the Government of India launched the landmark National Agroforestry Policy in 2014 (http://www.indiaenvironmentportal.org.in/content/389156/national-agroforestry-policy-2014/), which aims to mainstream tree growing on farms and meet a wide range of developmental and environmental goals.

Vegetation Carbon Sequestration Potential of AFS in India

Agroforestry systems, which occur under diverse ecological conditions in India, offer immense scope for enhancing carbon stocks in the terrestrial ecosystems. During photosynthesis, atmospheric CO2 is fixed as C in vegetation, detritus, and soil pools for “secure” storage. Vegetation carbon pools include those long-lasting products derived from biomass such as timber and belowground biomass such as roots. Nair et al. (2009a, 2010) reviewed the global literature on CSP of AFS and highlighted that aboveground CSP of AFS is tremendously variable, ranging from 0.29 to 15.21 Mg ha−1 year−1. Dhyani et al. (2016) reviewed the Indian literature on this topic and found that the CSP values (aboveground) range from 0.25 to 19.14 Mg C ha−1 year−1 for the tree components; and for bamboo-based systems, it may be as high as 21.36 Mg C ha−1 year−1 (Nath and Das 2012). A perusal of the data in Table 2, which summarizes the relatively recent studies on this, echoes the gross variability in CSP values of Indian AFS: aboveground C sequestration ranges from 0.23 to 23.55 Mg C ha−1 year−1 and belowground (root) C sequestration varies from 0.03 to 5.08 Mg C ha−1 year−1. Given the diverse nature of tree components involved, besides variations in ecoclimatic conditions, site quality, and stand management practices adopted, this is not unusual. The following section provides a brief account of the major factors influencing aboveground CSP of AFS.

Table 2 Biomass (aboveground + roots) carbon sequestration potentials of some agroforestry systems in India
Full size table

Agroforestry Systems and the Nature of Components

As mentioned, the diverse range of ecoclimatic conditions and the disparate array of agroforestry systems and practices in India representing profound variability in species and management regimes result in enormous variability of CSP values. In general, woodlots of bamboos, Acacia auriculiformis, A. mangium, and Populus deltoides are characterized by relatively high CSP (Table 2). Likewise, boundary plantation of 8-year-old P. deltoides had lower carbon stocks (4.51 Mg ha−1) than block plantations (28.67 Mg ha−1) in the Central Himalayan region (Kanime et al. 2013) with carbon sequestration rates of 0.43 and 2.75 Mg C ha−1 year−1, respectively. Mangalassery et al. (2014) found that silvopastoral systems involving Acacia tortilis and Azadirachta indica and grasses such as Cenchrus ciliaris and C. setigerus showed higher sequestration potential compared with systems containing only trees or pastures in the arid northwestern India.

While most AFS (e.g., multipurpose trees, silvopasture, energy plantations) have great potential for C sequestration, homegardens are unique in this respect. They not only sequester C in biomass and soil, but also conserve agrobiodiversity (Kumar 2006). Tilman et al. (1997) and Kirby and Potvin (2007) have suggested that plant assemblages with high species diversity may promote more efficient use of site resources compared with those of lesser diversity. It signifies that “biodiverse” systems such as tropical homegardens can maintain greater net primary production and consequently higher CSPs than AFS with fewer species. In a case study from peninsular Indian homegardens, Kumar (2011) found that average aboveground standing stock of C ranged from 16 to 36 Mg ha−1. Structural attributes such as size of the homegardens, however, may alter the carbon sequestration rates; for example, small homegardens in the reported study showed higher C stocks on unit area basis than large- and medium-sized ones.

Ecoregions and Site Quality

Agroforestry systems on humid and tropical sites have higher potential to sequester carbon than those on arid, semiarid, and temperate sites. For example, AFS in the Western Himalayan and humid tropical regions showed higher CSP than those in the arid and semiarid regions (Table 2). Ajit et al. (2017a) using the dynamic carbon accounting model, CO2FIXv3.1, simulated the CSP of extant AFS in 26 districts of 10 selected states in India over a 30-year period. Comparisons across districts indicate that CSP ranged from 0.05 to 1.03 Mg C ha−1 year−1 with a mean value of 0.21 Mg C ha−1 year−1. In another study involving the CO2FIX model, these authors (Ajit et al. 2017b) showed that the CSP (tree, crop, and soil) of the extant AFS in Kupwara district of Kashmir valley involving species such as Malus (33.75%), Populus (29.91%), Salix (14.32%), Juglans (6.68%), and Robinia (4.7%) was 0.88 Mg C ha−1 year−1. The CSP of an AFS, apart from the nature of the species involved (section “Species and Stand Age”), is driven by stand management (section “Silvicultural Management”) and the prevailing ecological quality of the site (site quality). In spite of the potential benefits of site-specific ecological conditions in enhancing stand growth, there are no studies addressing the impacts of site quality on CSP of AFS.

Altitudinal ranges as reported by some authors significantly influence carbon density (amount of carbon per unit area for a given ecosystem or vegetation type). For example, Rajput et al. (2015) showed that biomass carbon density in Kullu valley (Northwestern Himalayas) increased from 1000 to 1600 m altitude and declined thereafter, presumably because of the lower cropping intensity and shorter growing period prevailing in the upper altitudinal zones, which depress carbon density. As a result, carbon stocks/density may decline in the aboveground biomass and woody debris at high elevations (>1600 m). However, the soil organic carbon (SOC) may increase with elevation, albeit modestly, owing to the lower organic matter decay rates prevailing at higher altitudes, offsetting any net change in total carbon density (vegetation + soil) with increasing elevation.

Species and Stand Age

Choice of species is an important criterion that determines the carbon stocks of AFS. Fast-growing species such as bamboos, acacia (A. mangium; A. auriculiformis), poplar, eucalypts, and leucaena are generally characterized by high CSPs (Table 2). Dhyani et al. (2016) also reported similar results. Russell and Kumar (2019) using the CENTURY model showed that inclusion of trees with traits that promoted C sequestration such as lignin content, along with the use of best management practices, resulted in higher biomass (and therefore higher CSP), suggesting that the nature of tree components, besides the tree and stand management practices, holds the key in this respect. While evaluating the carbon sequestration in an age series of P. deltoides, a short-rotation plantation crop in Tarai region of central Himalaya, Arora et al. (2014) found that the C sequestration rate (in wood products and by substitution of biomass for coal) in mature plantations (7–11 years) varied from 5.8 to 6.5 Mg C ha−1 year−1. They also showed that aboveground carbon stocks increased from 0.5 Mg ha−1 in 1-year-old stands to 90.1 Mg ha−1 at 11 years of age, implying the dominant role of stand age in determining carbon stocks. Due to fast growth rate and adaptability to a range of environments, short-rotation plantations, in addition to high carbon storage, produce biomass for energy and contribute to reduced greenhouse gas emissions (Kaul et al. 2010). They also reported that high net annual carbon sequestration rates were achieved for fast-growing short-rotation poplar (8 Mg C ha−1 year−1) and eucalyptus (6 Mg C ha−1 year−1) plantations compared to the moderately fast-growing teak (Tectona grandis; 2 Mg C ha−1 year−1) and the relatively slow-growing (long-rotation) sal (Shorea robusta) forests (1 Mg C ha−1 year−1).

Silvicultural Management

Carbon sequestration being a function of tree growth and productivity, stand management practices (stand density regulation through thinning or through controlling initial planting density, pruning, fertilization, and weeding), apart from increasing the quality and quantity of production, may also promote C sequestration. In general, fast-growing tropical conifers and broad-leaved species respond favorably to silvicultural treatments. Information on the effect of planting density, crown pruning, and other management practices on the C accumulation potential, however, is scarce in the Indian context. In one such study, Kunhamu et al. (2011) found that biomass C stock of A. mangium trees was significantly altered by planting density and pruning treatments. The total tree (aboveground + roots) C sequestration was higher for the 5000 trees ha−1 treatment (81.82 Mg ha−1) than that for the 625 trees ha−1 (41.39 Mg ha−1) at 6.5 years of age. Rocha et al. (2017) using the same experimental stand reported that CSP ranged from 5.55 to 12.68 Mg ha−1 year−1 at 12 years of age with denser stocks having substantially higher values (Table 2). In another study involving a 30-year-old Hardwickia binata-based AFS in the hot semiarid environment of Rajasthan, Gupta et al. (2019) also reported a significant impact of tree population density on carbon sequestration. Average biomass carbon sequestered per tree (118.44 ± 50.26 kg C tree−1) was significantly more (44.5%) in the low-density (333 tree ha−1) stand compared to the high-density (666 tree ha−1) system. However, the total biomass carbon sequestered per hectare was significantly more (40.8%) in the high-density stand (31.6 ± 12.6 Mg C ha−1), implying the silvicultural trade-off between maximization of individual tree growth and maximization of stand growth.

Soil Carbon Sequestration

Soil carbon pool refers to the relatively stable forms of organic and inorganic C in the soil, which account for about two-thirds of the total C sequestration. Biomass such as plant residues that is not removed from the site is eventually incorporated into the soil as soil organic matter (SOM). Apart from plant residues, tree roots (both coarse roots and fine roots), which represent about one-fifth to one-fourth of the total living biomass, signify another important input of organic matter into the soil. SOM plays a vital role in determining C storage in terrestrial ecosystems and in regulating atmospheric CO2 fluxes. Soil C sequestration (SCS), therefore, is a significant greenhouse gas removal strategy (Lal 2008). However, literature on SCS potential of AFS in India, as it is generally the case elsewhere, is very scanty. Yet another problem is that many of the reported studies lack the required rigor (e.g., low sampling intensity, inadequate sampling depth, and/or inappropriate analytical procedures employed: section “Measurement and Estimation of C Sequestration in Agroforestry Systems”), making generalizations somewhat difficult.

Reviewing the global literature on SCS in AFS, Nair et al. (2009a) reported that the estimates vary greatly across systems, ecological regions, and soil types. The “best-bet estimates” ranged from 5–10 kg C ha−1 in about 25 years in extensive tree-intercropping systems on arid and semiarid lands to 100–250 kg C ha−1 in about 10 years in species-intensive multistrata shaded perennial systems and homegardens of the humid tropics (Nair et al. 2009b). In the Indian context, soil carbon stocks in AFS (0–100 cm depth) varied from 10.02 Mg C ha−1 for Ziziphus mauritiana + grass system in the arid western Rajasthan to as high as 229.5 Mg C ha−1 in the homegarden systems of Mizoram (Table 3). Like vegetation carbon stocks (Table 2), SCS potential was relatively low for the AFS in the arid and semiarid ecosystems compared to that of the humid tropical ecosystems (e.g., homegardens and woodlots; Table 3), which is consistent with the global trends mentioned above. Indeed, Saha et al. (2010) reported that soil carbon stocks of multistrata homegardens in central Kerala were next only to the adjacent tropical moist deciduous forest ecosystems. Despite the generally low SCS potential of the arid northwest Indian ecosystems, silvopastoral systems were found to be promising. For example, Mangalassery et al. (2014) reported that the SOC and net carbon sequestered were greater in the silvopastoral system in the arid parts of Gujarat, which had 36.3–60.0% more total SOC stock compared to the tree system and 27.1–70.8% more SOC than the pasture system.

Table 3 Recent reports on soil carbon stocks of agroforestry systems in India
Full size table

The influence of AFS on SCS generally depends on the quantity and quality of biomass inputs provided by the tree and non-tree components of the system, besides soil attributes such as soil structure and aggregation. Taxa of the multipurpose tree (MPT), stand age, and stand density are key factors in this regard. Dhyani et al. (2020) reported that MPTs like Alnus nepalensis, Parkia roxburghii, Michelia oblonga, Pinus kesiya, and Gmelina arborea with high ground surface cover, constant leaf litterfall, and extensive root systems have huge potential for augmenting SOC levels and for enhancing soil aggregate stability. Silvicultural management of stands may also increase SOM prompting improved productivity, besides providing climate change mitigation effects—signifying a win-win situation. Very little, however, is known about the changes in soil C storage of MPT stands under differing stand density management regimes. In a solitary study, Kunhamu et al. (2011) reported that high stand densities (5000 and 2500 trees ha−1) promoted SCS in 6.5-year-old A. mangium stands (31.79 and 34.64 Mg C ha−1, respectively) in the top (0–15 cm) layer of the soil profile. Intense pruning (up to 50% of tree height), however, depressed overall tree growth and soil C stocks at high (5000 tree ha−1) and low (625 tree ha−1) stand densities, while at intermediate densities (2500 and 1250 tree ha−1), pruning exerted a beneficial effect, signifying the need to maintain optimal stand densities, besides adopting appropriate tree management practices, for reaping carbon sequestration benefits.

The association between biodiversity (especially plant diversity) and SCS has become a topic of considerable scientific interest. Saha et al. (2009) reported that the soil C stock was directly related to plant diversity of homegardens. They found that homegardens with higher species richness and tree density than monocultural systems had greater soil carbon stocks, especially in the top 50 cm of soil. Overall, within the 1 m profile, soil C content ranged from 101.5 to 127.4 Mg ha−1. Furthermore, small-sized gardens (<0.4 ha) that had higher tree density and plant species diversity had relatively more soil C per unit area (119.3 Mg ha−1) than large-sized (>0.4 ha) gardens (108.2 Mg ha−1).

Higher species richness of tropical homegardens may also ensure greater stability of the SOM fractions, especially at lower soil depths. Undeniably, SOM represents a significant carbon store and can remain in the soil for extended periods as a part of soil aggregates. The recalcitrant fraction of SOM is “protected” from further rapid decomposition by biochemical recalcitrance, chemical stabilization, and physical protection (Christensen 1996; von Luetzow et al. 2008). Biochemical recalcitrance occurs when the chemical composition of SOM involves aromatic polymers and other structures that are difficult for microbes to break down (Christensen 1996). A familiar example is lignin, one of the main constituents of woody plants. Russell and Kumar (2019) in the modeling study mentioned earlier indicated that inclusion of trees with traits that promoted C sequestration such as lignin, along with the use of best management practices, resulted in higher soil C storage. Studies on aspects of SCS and factors leading to aggregate formation and stability are scarce in the Indian context.

Measurement and Estimation of C Sequestration in Agroforestry Systems

Yet another factor that determines the magnitude of soil and vegetation carbon sequestration is the methods employed for estimating vegetation CSP and SCS. Biomass is often taken as a surrogate of total C and the aboveground CSP values are typically the direct spin-offs of biomass measurements made either through destructive procedures or by employing allometric equations (Table 2). To derive carbon stocks, the amount of harvested and standing biomass is summed up assuming that 50% of the biomass comprises C, which however is variable depending on tissue types. Whole-tree harvest procedures for biomass estimation are also cumbersome. General allometric equations (Brown 1997; Piccard et al. 2012; Chave et al. 2014) are, therefore, widely employed in forestry, and are recommended by UNFCCC (2006) for tree biomass estimation in AFS also. Biomass estimation equations, however, vary with species, age, bole shape, and/or bole wood density. This has created the dilemma of whether to use the generalized equation for tree biomass estimation in AFS or not. Clearly, there is a need to develop a robust generic allometry that accounts for the heterogeneity of tree diversity throughout the landscape (Kuyah et al. 2012a).

As mentioned, often equations built for predicting biomass of forest trees are used in AFS. Variations in tree management, however, can be a concern, which limit the use of standard allometric equations developed for forests in agroforestry; for instance, trees in AFS may be pruned depending on management objectives or may have different growth forms due to differences in spacing compared to natural (forest) systems (Nair et al. 2009a). The determination of biomass production from AFS, therefore, is a challenging task and makes extrapolation from one system to others difficult and sometimes unrealistic (Nair 2012). Biomass regression equations, generalized for a geographic region, have been developed in a few cases to minimize errors in estimated biomass that result from such variability in sampled trees (e.g., Kumar et al. 1998). However, such location-specific allometric equations are not available for many agroforestry tree species.

In addition to aboveground biomass fractions, belowground net primary productivity (biomass) is a major pool of C. However, belowground biomass is difficult to measure and only very few Indian studies have characterized that. Root-to-shoot ratio is commonly used to estimate belowground living biomass. The ratios, however, differ substantially among species and across ecological regions, posing a serious problem in estimating belowground C sequestration in living biomass. Allometric equations for predicting root biomass have been constructed internationally (e.g., Kuyah et al. 2012b), but they are yet to gain popularity.

Apart from the root biomass, organic C occurs in soils as microbial biomass, and as SOM in labile and recalcitrant forms. The intricate interactions among these different forms make the measurement of SCS also a formidable task. The Walkley-Black (WB) procedure (Walkley and Black 1934) has been parsimoniously employed for SOC determination in India and elsewhere; it involves digestion of organic matter in the sample through oxidation with potassium dichromate. Although fast, convenient, and inexpensive, it is semiquantitative in nature and does not completely recover the organic carbon in soil (Abraham 2013). In fact, complete oxidation of SOC does not take place and variable levels of carbon recoveries have been reported (e.g., 60–86%: Nelson and Sommers 1996), implying that underestimation of SOC is in the WB procedure. The problem of incomplete digestion of the organic matter in the WB method, however, has been partially resolved by supplying external heat during sample digestion in the modified WB protocol (Nelson and Sommers 1996). Dry combustion methods, widely used for routine laboratory analysis, are considered to be the “gold standard” and superior to wet digestion (Nayak et al. 2019). Spectroscopic techniques for sensing of SOC are also evolving rapidly; nevertheless, the conventional methods will continue to be used in the near future despite their limitations (Nayak et al. 2019). Another major issue is the lack of uniformity in soil sampling, especially the depth of sampling (see Table 3). Although this problem is universal in nature (Nair 2012), it is more acute in the Indian context. Most soil studies are restricted to the surface soil layers, i.e., to 20 or 30 cm depth. In view of the fact that tree roots extend to deeper soil horizons, and the role of subsoil in long-term stabilization of C, the need for sampling the deeper layers of the soil profile cannot be overemphasized. Overall, a uniform set of methods and procedures are not available for estimating C sequestration in AFS. Wide variations also exist in the procedures used for soil sampling and analysis, which can greatly affect the conclusions made when comparing the differences under various management practices, soils, environments, and social conditions (Nair 2012).

Concluding Remarks

Agroforestry systems abound in India with profound variability in the nature of components and their dynamics. Biological carbon sequestration (in vegetation and soil) is an intrinsic feature of agroforestry. Being a low-cost strategy, it has immense scope in the national climate change mitigation debate. In general, AFS with multistrata canopy architecture are characterized by higher CSP (aboveground) than those with simpler canopy structures. Likewise, AFS in the humid regions have higher aboveground CSPs than those in the arid and semiarid regions. Aboveground CSP values of Indian AFS reported in the literature range from 0.23 to 23.55 Mg C ha−1 year−1. More than half of the C assimilated is also transported belowground via root growth and organic matter turnover processes (e.g., fine root dynamics, rhizodeposition, and litter dynamics), which enrich the soil organic carbon pool. Species diversity (especially plant diversity), stand age, and stocking levels, besides depth of sampling, are key determinants of SCS. Soil carbon stocks (0–100 cm depth) varied from 10.0 Mg C ha−1 to as high as 229.5 Mg C ha−1, signifying great variability in SCS among the various ecoregions and AFS of India. Older, densely stocked (e.g., block plantations) and biodiverse AFS (e.g., multistrata homegardens) are more efficient in SCS. Much like the aboveground CSP, AFS in the arid and semiarid regions showed much less potential for SCS than those in the humid regions. Proper choice of AFS involving rapidly growing multipurpose tree species and adopting appropriate stand management practices are, therefore, key to enhancing the prospects of biological carbon sequestration and evolving national climate change mitigation strategies, which are cost effective.