Elsevier

Geoderma

Volume 161, Issues 3–4, 15 March 2011, Pages 182-193
Geoderma

Organic matter stabilization in soil aggregates: Understanding the biogeochemical mechanisms that determine the fate of carbon inputs in soils

https://doi.org/10.1016/j.geoderma.2010.12.017Get rights and content

Abstract

We studied the biochemical and biophysical processes of carbon sequestration in an intensive agroforestry system on two soils (Feralsol – Luero; Arenosol – Teso) in W. Kenya to elucidate the mechanisms associated with long-term carbon storage. Specifically, we looked at a top-down model (macro-aggregates form around organic matter particles and micro-aggregates form within the macro-aggregates) and a bottom-up model (micro-aggregates form independently and are incorporated into macro-aggregates) of soil aggregate formation. Soil samples were collected from experiments on improved tree fallows using different species and two tillage treatments; water-stable aggregates were extracted and sorted into three size classes: macro-aggregates (> 212 μm), meso-aggregates (53–212 μm) and micro-aggregates (20–53 μm). Organic matter characterization of each fraction was based on 13C isotope abundance, Fourier transform infrared (FTIR) spectroscopy and the abundance of polysaccharides. Improved fallows increased soil C by 0.28 and 0.26 kg m­2 in the top 20 cm of the soil profile in Luero and Teso, respectively. Tillage altered the distribution of aggregates among size classes. Changes in the δ13C signature in each fraction indicated that more of the new carbon was found in the macro-aggregates (35–70%) and meso-aggregates (18–49%) in Luero and less (9–17%) was found in the micro-aggregates. In Teso, about 40–80% of the new aggregate C was found in the meso-aggregates, 14–45% was found in the micro-aggregates and only 4–26% was found in the macro-aggregates. The meso-aggregates and macro-aggregates to a lesser extent, in both sites, were enriched in carboxylic-C and aromatic-C, indicating the importance of OM decomposition and plant-derived C in the stabilization of larger aggregates, supporting the top-down model of aggregate formation. Microbially derived polysaccharides play a leading role in the formation of stable micro-aggregates and carboxylic-C promotes stabilization through surface occlusion. This bottom-up process is essential to promote long-term carbon sequestration in soils. Additionally, the micro-aggregates at both sites were enriched in polysaccharides and had elevated ratios of galactose + mannose:arabinose + xylose than the other aggregate fractions, indicating the importance of microbial processes in the formation of stable micro-aggregates and supporting the bottom-up model.

Research Highlights

► Improved fallows increased soil C by 0.28 to 0.26 kg m­2 in the top 20 cm of the soil. > Most new carbon was found in the macro- and meso-aggregates. > Large aggregates were stabilized with plant-derived C. > Microbially derived polysaccharides were important in the stabilization of micro-aggregates.

Introduction

Since the adoption of the Marrakech Accords in 2002 the international policy-makers have refused to consider carbon sequestration in soils as a means of mitigating climate change, for various reasons (Schulze et al., 2002, Lövbrand, 2009). With the decisions taken at the recent meeting of the United Nations Framework Convention on Climate Change in Copenhagen, agriculture is back on the table and there is renewed interest in the opportunities to sequester carbon in agricultural soils (UNFCCC, 2010). Yet, as a scientific community, we have not fully resolved the issue of the potential for soil sequestration of atmospheric CO2 and the conditions necessary for successful long-term carbon storage. For example, studies looking at surface soils have concluded that reducing tillage increases soil C (Kern and Johnson, 1993, Bayer et al., 2000, Machado and Silva, 2001, Tan and Lal, 2005). More recently, several studies have shown that reducing tillage in the mid-western portion of the North American continent and in the Brazilian cerrado simply results in a redistribution of carbon (C) within the soil profile and that there is likely a small net loss of C over the entire soil profile (Baker et al., 2007, Jantalia et al., 2007, de Moraes Sá et al., 2009). Other work in Brazil (Sisti et al., 2004, Denf et al., 2007) suggests that the introduction of a legume into the rotation alters the biogeochemical cycling of C and results in organic matter accumulation.

At the process level, we know that SOM stabilization in soil aggregates is the principal mechanism for long-term sequestration of C in soil organic matter (SOM). Most studies divide soil aggregates into three classes based on their size and there is some variability across studies in the size classes used. While increases in SOM are generally associated with increases in C-rich macro-aggregates, long-term sequestration depends on stabilization of carbon in micro-aggregates (Tisdall and Oades, 1982, Six et al., 2000). However, the genesis and dynamics of these micro-aggregates remains controversial. Several authors have proposed a model where stable micro-aggregates are formed within macro-aggregates and suggested that management practices that reduce macro-aggregate turnover (e.g. reduced tillage) promote the formation of very stable micro-aggregates that ensure long-term sequestration (Six et al., 2000, Christensen, 1996; Pujet et al., 1995). The model suggests that physical occlusion of the micro-aggregates leads to greater protection from microbial breakdown and ultimately to long-term carbon sequestration. Macro-aggregates form initially through the encapsulation of organic matter by soil minerals. For simplicity, we will refer to this model as the top-down model of aggregate formation. An alternative model suggests a bottom-up process whereby micro-aggregates form through the interaction between mineral surfaces and organic matter with little protection in early stages of micro-aggregate formation (Emerson, 1959, Tisdall and Oades, 1982, Lehmann et al., 2007). These micro-aggregates are later incorporated into macro-aggregates as they form through the occlusion of plant-derived organic matter. The recent work by Lehmann et al. (2007) further points to the importance of microbial processes in the formation of stable micro-aggregates and ultimately in the long-term sequestration of carbon in soils.

To examine the interaction of these biochemical and biophysical processes, we chose to work with an intensive fallow agroforestry system as a model in which to study the processes leading to soil C sequestration. Agroforestry systems have the potential to sequester significant amounts of carbon and contribute to climate change mitigation (Buresh and Niang, 1997, IPCC, 2000, Albrecht and Kandji, 2003). Greater attention has been paid to sequestration in the tree components of these systems, but soil sequestration may be significant in many of these systems. We also chose to conduct this work in a sub-humid tropical highland site on soils that are moderately weathered to look at a situation that favoured biological processes in soil aggregation (Oades and Waters, 1991).

The intensive agroforestry system used in this study has been shown to improve crop production and increase soil carbon in a relatively short period of time (Buresh and Niang, 1997, Albrecht and Kandji, 2003, Mutuo P.K., 2004, Mutuo et al., 2005). Improved fallows follow an alternation between cereal crops and tree-legume fallow. The duration of trees in the cycle depends upon the level of soil degradation and the nature of the rainfall. Soil C storage in these systems represents the potential for long-term C storage, as long as trees remain in the rotation, but the storage capacity is largely dependent upon soil texture and total rainfall (Albrecht and Kandji, 2003). These systems have been widely used to improve a number of other conditions in agroecosystems such as restoring soil fertility and erosion control (Valentin et al., 2004, Mafongoya et al., 2006).

The aim of this study is to determine the fate of C inputs in the soil. Over the course of these experiments, large quantities of N-rich organic matter have been incorporated. We will look at several hypotheses: i) these organic matter inputs will significantly increase the proportion of aggregates of all sizes in the surface soil; ii) organic materials from the decomposing tree litter will bind these macro- and meso-aggregates; iii) that macro- and meso-aggregates will have significant plant-derived C functional groups as a result of recent inputs from tree litter; iv) micro-aggregates are formed by microbial processes and thus will have higher polysaccharide and aliphatic-C (microbially derived) contents; and v) carbohydrates in micro-aggregates will be largely derived from microbial decomposition of organic matter. Quantifying the importance of the soil sink, understanding the fate of C inputs, and elucidating the mechanisms that result in C stabilization in soils will contribute to our understanding of the potential for these systems to contribute to long-term C storage.

Section snippets

Description of sites

The highlands of western Kenya cover an area of about 85,000 km2, which represents 15% of the total land area of the country and accommodates about 40–50% of the total population. The area receives bimodal rainfall, with long rains occurring from March to June and the short rains from August to November, totalling 1500–1800 mm annually. Long-term mean temperature ranges between 22 and 24 °C. This study was part of a large project “Improved fallows by legume plants in eastern and southern Africa”

Aggregate size distribution

Fallow treatments did not have a strong effect on the amount of aggregates in the soil, although we note that a large portion of the soil in each sample was already aggregated (Table 3). Due to lab error, samples for C. grahamiana, C. paulina and T. vogelii at 5–10 cm and 10–20 cm depths were lost. The data for samples from the other treatments showed a slight trend towards an increase in micro-aggregates with depth at both sites (P = 0.070 and 0.061 at Luero and Teso, respectively), so we did not

Carbon sequestration

Agroforestry systems are good model systems for looking at mechanisms for soil C sequestration. Albrecht and Kandji (2003) summarized sequestration rates in improved fallows grown in East and West Africa and found rates varying between 0.7 and 2.5 Mg ha­1 y­1. They also showed that soil texture is an important factor that determines the rates of sequestration and part of the evidence provided came from the experiments considered in this paper. Increasing C stocks over a short period of time is

Conclusion

The rates of carbon accumulation in the soil in these highland sites of western Kenya were at the low end of IPCC estimates (IPCC, 2000) for humid tropical areas and about half of the rates estimated by Albrecht and Kandji (2003). Only a small fraction of the added carbon could be found in the micro-aggregate long-term storage pool. Most of the newly added carbon ended up in the coarser aggregates and is subject to turnover and loss in the event that OM inputs decline.

Plant-derived OM was

Acknowledgements

We thank Margaret Thiong'o and Edith Anyango for HPLC analyses and data analysis; Paul Mutuo for soil fractionation; Joash Mango and Donald Agwa for field work and sample collection; Elvis Weullow and Andrew Sila for infrared spectroscopy technical and data analysis support; Mercy Nyambura and the staff of the ICRAF Plant and Soil Analysis Laboratory for soil analyses; David Harris at UC Davis for isotope analyses. This work was generously supported by a grant from the European Union Inco-Dev

References (50)

  • C.P.J. Sisti et al.

    Change in carbon and nitrogen stocks in soil under 13years of conventional or zero tillage in southern Brazil

    Soil Tillage Res.

    (2004)
  • J. Six et al.

    Soil macroaggregate turnover and microaggregate formation: a mechanism for C sequestration under notillage agriculture

    Soil Biol. Biochem.

    (2000)
  • Z.X. Tan et al.

    Carbon sequestration potential estimates with changes in land use and tillage practice in Ohio USA

    Agric. Ecosyst. Environ.

    (2005)
  • C. Valentin et al.

    Responses of soil crusting, runoff and erosion to fallowing in the sub-humid and semi-arid regions of West Africa

    Agric. Ecosyst. Environ.

    (2004)
  • W. Amelung et al.

    Neutral and acidic sugars in particle-size fractions as influenced by climate

    Soil Sci. Soc. Am. J.

    (1999)
  • J.R. Buresh et al.

    Tithonia diversifolia as a green manure: awareness, expectations and realities

    Agroforest. Forum

    (1997)
  • M.V. Cheshire
  • Christensen, B.T., 1996. Evaluation of soil organic matter models. In: Powlson, D.S., Smith, P., Smith, J.U (Eds.),...
  • J.C. de Moraes Sá et al.

    Soil organic carbon and fertility interactions affected by a tillage chronosequence in a Brazilian Oxisol

    Soil Tillage Res.

    (2009)
  • K. Denf et al.

    Microaggregate-associated carbon as a diagnostic fraction for management-induced changes in soil organic carbon in two oxisols, Soil Biol

    Biochem.

    (2007)
  • R.H. Ellerbrock et al.

    Functional analysis of soil organic matter as affected by long-term manorial treatment

    Eur. J. Soil Sci.

    (1999)
  • W.W. Emerson

    Stability of soil crumbs

    Nature

    (1959)
  • C. Feller et al.

    Soil carbon sequestration in tropical areas: general considerations and analysis of some edaphic determinants for Lesser Antilles soils

    Nutr. Cycl. Agroecosyst.

    (2001)
  • E.G. Gregorich et al.

    Carbon mineralization in soil size-fractions after various amounts of aggregate disruption

    J. Soil Sci.

    (1989)
  • G. Guggenberger et al.

    Land-use effects on the composition of organic matter in particle-size separates of soil: I. Lignin and carbohydrate signature

    Eur. J. Soil Sci.

    (1994)
  • Cited by (143)

    View all citing articles on Scopus
    View full text