Literature Review
An Overview of Global Carbon
Understanding the concept of carbon (C) sequestration seems to be the most reasonable point to start a lesson, but C sequestration which is an aspect of C management cannot be absolutely understood and appreciated as one of the identified mitigation options against global warming and climatic change, without a critical survey of global C as documented by most researchers.
Three main reservoirs regulate the C cycle on earth (IPCC, 1990): the oceans ≈ 39000 x 1015 g (or Pg) of C; the atmosphere (≈ 750 Pg C), and terrestrial systems (≈ 2200 Pg C). The fourth reservoir which is a permanent sink − the geological reservoir, is estimated at 65.5 x 106 Pg (Kempe, 1979). Table 1 shows the principal global C pools comprising of oceanic, geologic, pedological, atmospheric, and biotic. These pools are interconnected with sizeable fluxes among them. For example, the atmospheric pool is increasing at the rate of 3.3 Gt C yr-1. The oceanic pool is absorbing about 92 Gt C yr-1 and emitting 90 Gt yr-1, with a net gain of 2 Gt yr-1. The biotic pool photosynthesizes 120 Gt C plant respiration and the remaining 60 Gt C yr-1 by soil respiration (Lal et al., 2007). Although the soil-vegetation C pool is small compared with that of the oceans, potentially it is much more labile in the short term (Batjes, 1996). On average, the soil contains about 2.5 times more organic carbon (OC) than the vegetation (≈ 650 Pg C) and about twice as much C as is present in the atmosphere (≈ 750 Pg C) (Batjes, 1998). The soil is the largest terrestrial pool of OC, with global estimates ranging from 1115 to 2200 Pg C (Batjes, 1992), 1576 Pg C (Eswaran et al., 1995), 1400 Pg C (Falloon et al., 1998) and 1220 Pg C (Sombroek et al., 1993). Estimates of global soil C content have also been made by several researchers including, Kimble et al. (1990), Buringh (1984), Bohn (1982), Batjes (1998), Lal (2002), Post et al.(1990), Johnson and Kerns (1991).
World soils or the pedological pool comprises two distinct components: soil organic carbon (SOC) and soil inorganic carbon (SIC) pools estimated at 1576 Gt and 938 Gt respectively, at 1 m depth (Post et al., 1982; Schlesinger, 1995; Eswaran et al., 1993). The SOC pool is concentrated in soils of arctic, boreal, and temperate regions which includes highly active humus and relatively inert charcoal C, while the SIC pool includes elemental C and carbonate minerals, such as calcite, dolomite, and gypsum and those of arid and semiarid climates (Schnitzer, 1991; Stevenson, 1994; Lal et al., 2000; Wagner, 1981; Paul and Clark, 1996). Lal (2004) recognized two types of carbonates in soils: primary or lithogenic carbonates and secondary or pedogenic carbonates. Of the global SOC pool (2500 Gt), which includes about 1550 Gt of SOC and 950 Gt of SIC therefore, the total soil C including both SOC and SIC pools in the active soil layer of 1 m depth constitutes about 23000 Pg (Lal, 2002b) which is 3.3 times the size of the atmospheric pool (760 Gt) and 4.5 times the size of the biotic pool (560 Gt) (Lal, 2004). Table 2 shows the global mass of SOC in the upper 30 cm, 1 m, 2 m, and 3 m of soil.
The SOC pool in the top 1 m depth of world soils ranges between 1462 and 1600 Pg, which is nearly three times that in the aboveground biomass and approximately double that in the atmosphere; 32 % (or 506 Pg) of this is contributed by soils in the tropics (Eswaran et al., 1993; Lal et al., 1995; Batjes 1996). According to Batjes (1996) total soil C pools for the entire land area of the world, excluding carbon held in the litter layer and charcoal, amounts to 2157 to 2293 Pg of C in the upper 100 cm. Owning to the problem of making accurate global estimates of C, Eswaran et al. (1993) suggested that employing the coefficient of variation (CV) which is an expression of the variability will aid in understanding and accepting the reliability of most generalization.
Lal et al. (2007) documented the estimates of global carbon pool up to 1 m depth of the various USDA soil orders. Another study that provided global mean of SOC estimates in soils of the tropics reported values of 8.3 for Ultisols, 9.7 for Oxisols, and 10.4 kg m-2 for Inceptisols among other soil orders (Lal, 2002a). These differences between soil orders in the tropics are mainly in relation to temperature, rainfall, soil texture and land use (Batjes, 2000). Similarly, Kimble et al. (1990) reported SOC density of principal soil orders of the world, where among others the mean SOC density is 9.7 kg m-2 with CV of 42 % for tropical Oxisols and 8.3 kg m-2 with CV of 70 % for tropical Ultisols. About 52 % of this C pool is held in the top 30 cm of the soil profile, the layer most susceptible to land use changes and responsive to management practices (Lal, 2002b).
2.2 Soil Organic Carbon Dynamics
Lal (2001) noted that the SOC pool which is a function of soil characteristics and climatic factors is a highly variable and dynamic entity. It is variable over space because its density differs widely among soils and ecoregions; and variable over time because it changes with change in land use and management. Lal (2002b) observed that SOC pool is in a dynamic equilibrium with its environment, with a balance of input and output at a steady state level. Thus, the SOC pool represents a dynamic equilibrium of gains and losses (Fig. 1). Follett (2001) pointed out that temporal variation in SOC level results from the balance between plant biomass input and decomposition and OC losses via leaching, oxidative, and erosional processes. Substrate quality is one of the main factors affecting decomposition and has been linked to the relative abundance of specific compounds such as nitrogen, lignin (Melillo et al., 1982; Tian et al., 1993), and phenolic acids (Martens, 2000). Lal (1999) hypothesized that lack of nutrients, especially N, could explain the low C conversion efficiency. Knops and Tilman (2000) observed
that the rate of carbon accumulation in agricultural abandoned fields was controlled by the rate of nitrogen accumulation, which in turn depended on atmospheric nitrogen deposition and symbiotic nitrogen fixation by legumes. More so, the turnover time of organic matter (OM) increases with depth in the soil, ranging from several years for litter to 15-40 years in the upper 10 cm and over 100 years below a depth of about 25 cm (Harrison et al., 1990; Lobo et al., 1990). In soils of the tropics, particle size fractionation techniques have been used to characterize relationships between SOC and aggregation at the macro and microaggregate scale (Feller et al., 1996). The concept is that soil organic fractions associated with different sized particles differ in structure and function, and therefore play different roles in SOC turnover (Christensen, 1992).
The SOC pool consists of “a mixture of plant and animal residues at various stages of decomposition, of substances synthesized microbiologically and/or chemically from the breakdown products, and of the bodies of live microorganisms and soil animals and their decomposing products” (Schnitzer, 1991). The different C pools existing in the soils have been described in terms of the different mean residence times, ranging from years (active fraction), to decade to hundreds of years (passive), to thousands of years (stable) (Carter et al., 2002). The C pools are relative concepts based on the rate of decomposition of particular constituents and are more related to biological function than to particular soil chemical C constituents. For example, the active fraction consists of live microorganisms (microbial biomass), microbial products, and unprotected chemical constituents such as proteins and polysaccharides with a turnover time of a few weeks or months.
The slow fractions are more resistant to decomposition due to partial physical and chemical protection with a longer turnover time (Theng et al., 1989). The passive organic constituents include humic substances and other macromolecules that are intrinsically resistant against microbial attack due to chemical recalcitrance, physical protection by adsorption on mineral surfaces, or entrapment within soil aggregates (Gregorich et al., 1997). Lal (2001) pointed out that formation of stable microaggregates in the subsoil takes C out of circulation by encapsulating it (physical and chemical protection from microbial activity) and is thus sequestered. The long-term stabilization of C in temperate and tropical soils is mediated by soil biota (e.g. fungi, bacteria, roots and earthworms), soil structure (e.g. aggregation) and their interactions, and is influenced by agricultural management (Six et al., 2002).
In most soils, C is organic and constituents approximately 57 % of the soil organic matter (SOM) that includes a wide spectrum of organic compounds, from labile components, such as relatively fresh plants material and microbial biomass, to refractory components such as charcoal, which accumulates slowly over thousands of years (Trumbore, 1993). Piccolo (1996) indicated that the process of turning agricultural soil into sink for OC sequestration would be complete if the stored OM were transformed into stable and recalcitrant humic substances. Accordingly, humified OC, humic acids and humin in particular, represent the most persistent pool for SOC
Fig. 1: Processes affecting soil organic carbon dynamics.
Arrows pointed upward indicate emissions of Co2 into the atmosphere (Adapted from Lal, 2004).
accumulation with mean residence time of several hundreds of years. Spaccini et al., (2002) is of the view that hydrophobic protection provided by humified matter may substantially reduce decomposition of labile organic compounds in soils, thus they reported that the higher the hydrophobicity of a humic material, the larger the sequestration of OC in soil. Bayer et al. (2000) reported that in southern Brazil, SOC associated with sand and silt fractions was less humified than that associated with finer-sized fractions. Nevertheless, Oades et al. (1987) and Six et al. (2000) demonstrated that the most humified or oldest fraction is associated with silt particles. Most of the C losses following soil disturbance such as tillage originate from the active and slow pools, which comprise the biologically defined SOM pools described as active (labile), slow (partially labile), and passive (stable) (Jenkinson and Rayner, 1977; Jenkinson, 1990; Duxbury and Nkambule, 1994). Biological separation of SOC empirically separates labile from recalcitrant forms by allowing microbes to mineralise C under controlled conditions with the most labile C mineralised first and recalcitrant C mineralised later.
Physical fractionation of the soil according to aggregate size has been used to study the portioning of OC in the soil (Buyanovsky et al., 1994). Fractionation using physically based models is possible because OC is protected within and between aggregates (Cambardella and Elliot, 1992; Six et al., 2000; Snyder and V´azquez, 2004). In its simplest case, there is a free light fraction (LF) of labile C between the aggregates and intra-aggregate particulate OM (iPOM) within macroaggregates (Cambardella and Elliot, 1992; Six et al., 1998). The LF may be more related to residue input rates and soil environmental conditions and the iPOM more related to aggregate turnover, which is strongly affected by tillage management (Six et al., 1998). Aggregate hierarchy levels of formation occur in which the intra-macroaggregate POM (particulate organic matter) facilitates the binding of microaggregates into macroaggregates, which in turn affects the variation in the accessibility of soil microorganisms to SOC that leads to pools which differ in stability and dynamics. For example, in relatively undisrupted systems such as no-tilled agricultural and native systems, the greatest C concentration is usually found in the small macroaggregate size class (250− 2000 μm), with C in this fraction being most affected by cultivation (Beare et al., 1994; Cambardella and Elliott 1994). Furthermore, a greater proportion of the SOC pool in large microaggregates implies greater C losses to the atmosphere if macroaggregates are broken by soil management practices (Cruz-Rodríguez, 2004). The losses of SOC are in turn associated with losses in the POM fraction and therefore the amount of aggregation and aggregate turnover (Six et al., 1999).
