Soil organic carbon fraction losses upon continuous

Geoderma 209–210 (2013) 214–225

Contents lists available at SciVerse ScienceDirect

Geoderma journal homepage: www.elsevier.com/locate/geoderma

Soil organic carbon fraction losses upon continuous plow-based tillage and its restoration by diverse biomass-C inputs under no-till in sub-tropical and tropical regions of Brazil Florent Tivet a,c,⁎, João Carlos de Moraes Sá b, Rattan Lal c, Paulo Rogério Borszowskei d, Clever Briedis e, Josiane Bürkner dos Santos e, Márcia Freire Machado Sá b, Daiani da Cruz Hartman f, Guilherme Eurich f, Anderson Farias f, Serge Bouzinac a, Lucien Séguy a a

Centre de Coopération Internationale en Recherche Agronomique pour le Développement, CIRAD, UPR SIA, F-34398 Montpellier, France Department of Soil Science and Agricultural Engineering of State University of Ponta Grossa, Av. Carlos Cavalcanti 4748, Campus de Uvaranas, 84030-900 Ponta Grossa, PR, Brazil c Carbon Management and Sequestration Center, School of Environment and Natural Resources, The Ohio State University, OARDC/FAES, 2021 Coffey Road, Columbus, OH 43210, USA d CESCAGE — Centro de Ensino Superior do Campos Gerais, Av. Carlos Cavalccanti 8000, Uvaranas, Ponta Grossa, PR, Brazil e Graduate Program in Agronomy of State University of Ponta Grossa, Av. Carlos Cavalcanti 4748, Campus de Uvaranas, 84030-900 Ponta Grossa, PR, Brazil f Undergraduate Program in Agronomy of State University of Ponta Grossa, Av. Carlos Cavalcanti 4748, Campus de Uvaranas, 84030-900 Ponta Grossa, PR, Brazil b

a r t i c l e

i n f o

Article history: Received 1 February 2012 Received in revised form 20 May 2013 Accepted 14 June 2013 Available online xxxx Keywords: Cerrado Tropics Soil C fractionation Land use change Soil resilience Multifunctionality of cover/relay crops

a b s t r a c t The conversion of native vegetation (NV) into agricultural land by clearing and tillage disrupts the soil structure, and depletes soil organic carbon (SOC) pool. The data on changes in SOC pools are needed to enhance scientific knowledge regarding the effects of land use and no-till (NT) systems on soil fertility, agronomic productivity, and soil C sink capacity. Thus, the objective of this study was to quantify changes in SOC fractions due to conversion of NV to agricultural land, and to assess the rate of recovery of SOC fractions and the resilience index of NT cropping systems under sub-tropical (Ponta Grossa/PR — PG) and tropical (Lucas do Rio Verde/MT — LRV) regions of Brazil. The conversion from CT to NT was 29 and 8 years at the PG and LRV sites, respectively. Five different fractions of SOC pools were extracted by chemical methods (i.e., C in the polysaccharides — CTPS, hot-water extractable C — HWEOC, chemically-stabilized organic C — CSOC), and physical fractionation (i.e., particulate organic C — POC, and mineral-associated organic C — MAOC). Land use change primarily altered the labile (HWEOC, TPS, and POC) and also some of the stable (MAOC) pools at both sites. The CSOC pool was almost constant throughout the soil profile and represented, across land uses, 7.2 g C kg−1 at the PG and 3.1 g C kg−1 at the LRV sites. At the PG site, the HWEOC and CTPS concentrations in the 0–5 cm depth decreased by 56% (1.21 g kg−1) and 45% (7.21 g kg−1) in CT soil, respectively. At the LRV site, concentrations of HWEOC and CTPS in the 0–5 cm depth decreased by 50% (0.4 g kg−1) and 42% (4.8 g kg− 1), respectively. In contrast, concentrations of HWEOC and CTPS fractions in soil under NT in the 0–20 cm depth were closer than those under NV, and exhibited a distinct gradient from surface to sub-soil layers. The adoption of CT reduced POC by 46% (4.7 Mg ha−1), and MAOC by 21% (15.1 Mg C ha−1) in the 0–20 cm depth at the PG site. Using CT for 23 years at the LRV site, decreased SOC fractions in the 0–20 cm depth at the rate of 0.25 and 0.34 Mg C ha−1 yr−1 for POC and MAOC, respectively. In contrast, adoption of intensive NT systems in tropical agro-ecoregions increased POC at the rate of 0.23 to 0.36 Mg C ha−1 yr−1, and MAOC by 0.52 and 0.70 Mg C ha−1 yr−1. An important effect to be emphasized is the possibility of recovering, at least partially, the SOC fractions by adopting high biomass-C inputs under NT management, and despite the fact that the experimental duration at the LRV site was only eight years. With a high and diversified input of biomass-C in intensive NT systems, higher resilience index was observed for CTPS, HWEOC, and MAOC. The variation in SOC among CT and NT systems was mainly attributed to the MAOC fraction, indicating that a significant proportion of that fraction is relatively labile, and that spatial inaccessibility of SOC plays a significant role in the restoration of SOC. © 2013 Elsevier B.V. All rights reserved.

Abbreviations: CA: conservation agriculture; CEC: cation exchange capacity; CN: carbon nitrogen; CSOC: chemically stabilized organic carbon; CTPS: carbon in total polysaccharides; CT: plow-based conventional tillage; Febrapdp: Brazilian federation of no-till; HWEOC: hot-water extractable organic carbon; LRV: Lucas do Rio Verde; MAOC: mineral-associated organic carbon; MT: minimum tillage; NT: no-till; NV: native vegetation; OM: organic matter; PG: Ponta Grossa; POC: particulate organic carbon; SOC: soil organic carbon; SOM: soil organic matter; TOC: total organic carbon; TPS: total polysaccharides. ⁎ Corresponding author at: Centre de Coopération Internationale en Recherche Agronomique pour le Développement, CIRAD, UPR SIA, F-34398 Montpellier, France. Tel.: +33 4 6761 5643; fax: +33 4 6761 7513. E-mail address: fl[email protected] (F. Tivet). 0016-7061/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.geoderma.2013.06.008

F. Tivet et al. / Geoderma 209–210 (2013) 214–225

1. Introduction Soils can be a sink or source of atmospheric CO2 depending on climate (i.e., temperature and precipitation), soil mineralogy, net primary production (NPP), land use and management. Land use change, conversion of native vegetation (NV) to agricultural land, exacerbates CO2 emissions through deforestation, biomass burning, and soil organic carbon (SOC) depletion by conventional plow-based tillage (CT). Both the role and magnitude of depletion of SOC are higher in soils of the tropics than those of the temperate regions due to a rapid turnover time of SOC (Jenkinson and Ayanaba, 1977). The latter is driven by a lower adsorptive capacity of the predominant minerals (1:1 clay, iron and aluminum oxides), weaker soil aggregation and lower protection of SOC against mineralization (Six et al., 2002; Zotarelli et al., 2005). The primary driving forces for decline of SOC under CT are the disruption of soil aggregates, and marked changes in soil environment (i.e., temperature, moisture, and oxygen) thus affecting microbial activity, and the attendant greater access of SOC to microbial processes. In addition, to accelerated soil erosion (Lal, 1976), the bare soil surface under CT is exposed to frequent wet–dry cycles enhancing the turnover rate of aggregates (Beare et al., 1994b). In contrast, adoption of no-till (NT) with no soil disturbance, and where inputs of plant residues (aboveground biomass and roots) and associated turnover of soil biological activity are enhanced, can increase the SOC pool and its stability by physical protection within stable aggregates (Beare et al., 1994a; Six et al., 2000). Recognizing the role of soils as a potential sink for atmospheric CO2 (Feller and Bernoux, 2008; Lal, 2004), there is a growing interest in adopting NT to store C in soils for off-setting anthropogenic emissions of CO2, but also to restore SOC pool lost by the conversion of NV and to improve and sustain productivity of agroecosystems. Several studies (i.e., Bayer et al., 2000; Paustian et al., 1992), have shown a strong relationship between input of the biomass C and the SOC pools. For a sandy clay loam Acrisol in southern Brazil, Bayer et al. (2000) reported that about 19% of C added in crop residues, without considering root biomass, became part of the soil as SOC pool. Several regions of Brazil (South and Cerrado) experienced severe soil degradation and SOC depletion because of intensive tillage, and high rate of SOC mineralization under tropical conditions (Bayer et al., 2004; Dieckow et al., 2009; Sá et al., 2001). The principal challenge under hot and humid Cerrado climate with a rapid turnover of soil organic matter (SOM) and nutrients was to develop high biomass-C inputs managed under NT. Séguy et al. (1998, 2006) tested a range of diversified NT mulch-based double cropping systems (i.e., a grain crop in summer and a cover crop – or combination of cover crops – established at the beginning and/or at the end of the rainy season) using several species (Pennisetum typhoides, Sorghum bicolor, Crotalaria spp., Eleusine coracana, Brachiaria spp., S. guianensis, etc.) well adapted to acidic soils and allowing a significant increase of biomass production. By 2010, NT systems are practiced on approximately 25.5 million ha of cropland in southern Brazil and the Cerrado region (Febrapdp, 2011). Numerous studies have been conducted to assess the changes in SOC pool following the conversion of NV into agricultural lands managed under CT, and upon conversion of CT to NT systems. Conversion to NT systems increases the SOC pool at least in the soil surface layer (Batlle-Bayer et al., 2010). However, most studies have assessed the total SOC pool, and the research information is scanty regarding changes in sensitive or labile SOC fractions, and C dynamics in relation to different stabilization processes under predominant Brazilian agro-ecoregions. It is widely reported that conversion of CT to NT increases particulate and (but not always) mineral-associated SOC fractions depending on the intensity of NT (i.e., biomass input), climate, duration of conversion from CT to NT, and clay mineralogy (Bayer et al., 2004; Sá et al., 2001). In addition to the total SOC pool, there also occur qualitative changes in SOC fractions under the NT systems. In general, the SOC recalcitrance decreases under NT systems, indicating that crop residues added are only partially decomposed, exceed microbial metabolic rate, and form less humified SOC fractions (Bayer et al., 2000; Tivet et al., 2013b).

215

A better understanding of the effects of land use change and agricultural management on long-term C sequestration necessitates separation of SOC into fractions. The SOC pool is highly diverse, composed of fractions with contrasting turnover times, and stabilized or protected against microbial decomposition through three main mechanisms: (i) spatial inaccessibility within aggregates, (ii) selective preservation through biochemical recalcitrance, and (iii) chemical protection through interactions with mineral surfaces (von Lutzow et al., 2006). However, the importance of selective preservation to explain the differences between the three main SOC pools (active, intermediate and passive) was largely overestimated, and that the two other mechanisms are preponderant for long-term SOC stabilization. A conceptual model with three principal types of SOC pools described by Sollins et al. (1996) and von Lutzow et al. (2008) indicates the following: (i) active or labile pool has a short residence time (i.e., a few months to a decade); is easily impacted by soil use and management; and is comprised of microbial biomass C (MBC), plant residues, roots, and fungal hyphae at different decomposition stages; and is stabilized by selective preservation; (ii) slow or intermediate pool has a half-life of 10 to 100 years; is comprised of decomposed residues, is stabilized by organic matter (OM) occlusion and through organo-mineral interactions; and is less impacted by land use and management; and (iii) passive pool has half-life of N 100 years; is comprised of stable or recalcitrant and mineral-associated organic C (MAOC), is stabilized through encapsulation within stable micro-aggregates, in clay micro-structures and interactions with minerals; and is least impacted by land use and management. The passive pool is rather heterogeneous in terms of the turnover time and composition, hence, represents a mixture of intermediate and passive SOC fractions. An array of physical and chemical (i.e., oxidation, hydrolysis) methods is used to isolate different functional fractions involved in diverse processes in soil including, among others, biological activity, C distribution and stabilization, and soil structural stability. Physical fractionation is a useful tool to provide a rough differentiation between active, intermediate and passive SOC pools, and also to assess the impact of soil management on dynamics (Cambardella and Elliott, 1994; Sá et al., 2001) and qualitative changes (Bayer et al., 2000). In addition, chemical methods are also used to isolate: (i) hot-water extractable organic C (HWEOC) representing the microbial C (Ghani et al., 2010) involved in aggregate stability (Haynes and Swift, 1990), (ii) acid hydrolysable total polysaccharides (TPS) which are the main binding agents involved in soil aggregation (Haynes and Francis, 1993; Tisdall and Oades, 1982), and (iii) a refractory C fraction obtained through oxidation with H2O2 which oxidizes labile organic compounds but retains chemically stabilized organic C (CSOC) (Jagadamma and Lal, 2010). A combination of physical and chemical methods can provide valuable information to fingerprint quantitative and qualitative changes in SOC pool due to land use and agricultural management. Sá et al. (2013) assessed changes in total SOC stocks upon conversion of NV to CT, and from CT to NT in sub-tropical and tropical regions of Brazil. Relatively rapid changes in SOC stock with intensive NT systems (e.g. diversity of cover/relay crops and high annual biomass input) indicate a large potential to reverse the process of soil degradation and of SOC depletion through conversion to NT systems based on complex rotations and cover cropping in the Cerrado region of Brazil (Séguy et al., 2006). The present study is based on the hypothesis that several SOC fractions are differentially impacted by land use changes, and that losses can be restored by the adoption of NT systems. It is also hypothesized that the magnitude of SOC fractions recovery is driven by the rate of biomass input. Thus, the specific objectives of this study were to assess: (i) the magnitude of changes in SOC fractions (i.e., HWEOC, TPS, CSOC, particulate and mineral-associated organic C) by conversion of the NV into agricultural land and CT, (ii) the spatial distribution of SOC fractions among land use and management practices, and (iii) the rate of recovery of SOC fractions and the resilience index (RI) of NT systems for different inputs of biomass C.

216

F. Tivet et al. / Geoderma 209–210 (2013) 214–225

2. Materials and methods 2.1. Site locations The characteristics of both sites (i.e., location, soil type, texture and mineral properties), details of cropping systems, as well as biomass input, are given by Sá et al. (2013). Briefly, field experiments were conducted (i) near the town of Ponta Grossa, in Paraná State, southern Brazil (25°09′S–50°09′W, 865 m above sea level), at the experimental station of IAPAR (Instituto Agronomico do Paraná), and hereafter called the PG site; and (ii) near the city of Lucas do Rio Verde, in Mato Grosso State, center western Brazil (13°00′S–55°58′S, 380 m above sea level), at the experimental station of Fundação Rio Verde, and hereafter called the LRV site. 2.2. Experimental design and land use management 2.2.1. Ponta Grossa site (PG) The climate is subtropical (Cfa, according to Koppen's classification), with mean annual average temperature of 18.5 °C, and mean maximum and minimum temperatures of 27.5 °C in February and 10.0 °C in July, respectively. Mean annual precipitation is 1545 mm, with January being the wettest month (185 mm) and August the driest (80 mm). The experiment is established on an Oxisol (Rhodic Hapludox, USA classification) derived from the clastic sediments of the Devonian period characterized by a mixture of Ponta Grossa shale (Soil Survey Staff, 1994). The experiment, established in 1981, comprises three tillage treatments: (i) conventional tillage (CT; C-input = 3.07 Mg ha−1 yr−1) such as a plow tillage after summer harvest and one after winter harvest and two narrow disking; (ii) minimum tillage (MT; C-input = 2.99 Mg ha−1 yr−1) comprising of one chisel plow and one narrow disking; and (iii) continuous no-till (NT; C-input = 4.15 Mg ha− 1 yr− 1) without any soil disturbance. The dimensions of each plot were 100 × 100 m for the NT treatment, and 50 × 140 m for the CT and MT. The cropping system for all tillage treatments comprised of a three year crop sequence with two crops per year with soybean (Glycine max, L. Merril) in six summers and maize in four summers in the last ten years alternating with oats (Avena strigosa Schreb), wheat (Triticum aestivum L.) and vetch (Vicia sativa) in the winter. In close proximity to the experimental plots, a NV plot was selected as a baseline for comparisons. Six sub-plots were marked on cropped fields and NV for soil sampling. The average concentration of Fe, Si and Al oxides did not vary among treatments or depth (Goncalves et al., 2008), and soil of all plots has a similar textural composition. 2.2.2. Lucas do Rio Verde (LRV) According to the Köppen classification, the local climate is Aw (Tropical Rainy) with rainfall concentrated in the summer (October through April). The site is characterized by a pronounced dry season between May and September. The mean annual rainfall is ~1950 mm with an average temperature of 25.2 °C. The dominant soil of the LRV site is a Red Latosol (Typic Haplustox, USA classification), derived from shale and sandstone parent materials (EMBRAPA, 1999). The experiment was established in 2001 on a total area of 10.9 ha. Seven tillage treatments have been implemented: (i) soybean alternating with cotton under conventional plow-based tillage designed as CT [C-input = 4.01 Mg ha−1 yr−1]; (ii) NT1 — soybean in the summer followed by maize (Zea mays L.) + ruzi grass (Brachiaria ruziziensis) as second crop [C-input = 7.60 Mg ha−1 yr−1]; NT2 — soybean in the summer followed by finger millet (Eleusine coracana) or finger millet + pigeon pea (Cajanus cajan) as second crop [C-input = 7.25 Mg ha− 1 yr− 1]; NT3 — soybean in the summer followed by finger millet + pigeon pea or finger millet + Crotalaria spectabilis as second crop [C-input = 6.84 Mg ha−1 yr−1]; (iii) NT4 — soybean in the summer followed by finger millet + C. spectabilis or sunflower

(Helianthus annuus) + B. ruziziensis as second crop [C-input = 7.34 Mg ha− 1 yr− 1]; (iv) NT5 — soybean in the summer followed of sorghum (Sorghum bicolor) + B. ruziziensis as second crop [C-input = 8.38 Mg ha−1 yr−1]; and (v) NT6 — soybean in the summer followed of millet (Pennisetum glaucum) or maize + B. ruziziensis as second crop [C-input = 7.41 Mg ha−1 yr−1]. The experience gained in the Cerrado region of Brazil by Séguy et al. (1998) was essential to build cropping sequences with diverse and high biomass-C inputs managed under NT. These NT systems are based on the insertion of supplementary biomass-generating crops, which are intercropped before or after the commercial crops or in relay, enhancing ecosystem services (Séguy et al., 2006). The dimension of each plot was 216 × 42 m for the NT treatments with three sub-plots (72 × 42 m), and 216 × 252 m for the CT treatment with four sub-plots (216 × 62 m). An adjacent NV Cerrado was included to represent the original vegetation and the undisturbed soil conditions as the baseline, and six sub-plots were demarcated for soil sampling. 2.3. Sampling Soil samples were obtained at the PG site in September 2009 after 29 years of cultivation, and at the LRV site in October 2009 after 8-years of cultivation. Soil samples were collected from: 0–5, 5–10, 10–20, 20–40, 40–60, 60–80, to 80–100 cm depths. Bulk soil samples were obtained for the 0–5, 5–10 and 10–20 cm depths by digging 15 × 15 cm trenches. Soil samples for the 20–40, 40–60, 60–80 and 80–100 cm depths were obtained with an auger (4.5 cm diameter). Samples obtained from four randomly selected points within each sub-plot were composited. A fifth central point was selected for measuring soil bulk density (ρb). Soil sampled from the NV plot was analyzed to establish the baseline. Soil samples were collected at the same depth intervals, and following the same procedure as described previously for the tillage treatments. Soil ρb was measured for each depth by the core method (Blake and Hartge, 1986) using 5 × 5 cm cores. The cores from 10–20 to 80–100 cm were taken in the middle section of the corresponding layer. The bulk samples were ovendried at 40 °C, gently ground, sieved through a 2-mm sieve, and homogenized. 2.4. Soil organic carbon pool extraction and analysis Details of the fractionation methods are described in the following section. Different SOC fractions were isolated by: (i) chemical methods to obtain the hot-water extractable organic C (HWEOC), the acid-extracted polysaccharides (TPS), and the chemically stabilized organic C (CSOC) extracted by H2O2 oxidation; and (ii) physical fractionation to separate the particulate organic carbon (POC), wet sieving of the whole sample isolating the particle size fraction between 53 and 2000 μm (POC) and the slow and passive pool of mineral-associated organic carbon (MAOC) which is b 53 μm. 2.4.1. Hot-water extractable organic carbon (HWEOC) The HWEOC was determined by the method of Ghani et al. (2003). Briefly, 3 g of bulk soil was weighed into 15 mL polypropylene centrifuge tubes. The sample was treated with 10 mL of distilled water for 16 h at 80 °C. Each tube was then shaken to ensure that the HWEOC released from the SOC was fully suspended in the solution. The tubes were centrifuged for 10 min at 4000 rpm. The SOC in the centrifuged extracts was oxidized by dichromate in sulfuric acid and back titrated with ferrous sulfate. The HWEOC is representative of the microbial biomass, containing more microbially derived than acid hydrolysable carbohydrates (Haynes and Francis, 1993). Given the strong correlations between HWEOC and other biochemical measurements (microbial biomass, mineralizable N), HWEOC could be used as an integrated measure of soil quality (Ghani et al., 2003).

F. Tivet et al. / Geoderma 209–210 (2013) 214–225

2.4.2. Acid-extracted total polysaccharides (TPS) Concentration of TPS was determined by the method of Lowe (1993). Briefly, 0.5 g of bulk soil was transferred to a 125-mL Erlenmeyer flask along with 4 mL of 12 M H2SO4. After 2 h, the H2SO4 was diluted to 0.5 M by adding 92 mL of distilled water. The flask was then placed on a water bath at 100 °C for 2.5 h. After cooling, the contents were filtered into a 125-mL volumetric flask. Distilled water was used to wash the residues to obtain 125 mL of filtrate. Of this, 1 mL was withdrawn with a pipette and transferred to a test tube, to which 1 mL of 5% (w/v) phenol was added followed by 5 mL of concentrated H2SO4 (96%, w/v). The tubes were allowed to stand for 10 min, and then placed in a water bath at 25–30 °C for 25 min. The absorbance of the solutions was read by a spectrophotometer at 490 nm. The conversion of the absorbance to TPS concentration (g kg−1) was done by using a calibration curve obtained with absorbance values of glucose standard. Concentration and stock of TPS, based on a mass ratio of 0.40 for C:CH2O in carbohydrates was used, and is hereafter called CTPS. The HWEOC and CTPS refer to labile organic C pools mostly involved in diverse processes in soil including the biological activity (Sparling et al., 1998), C distribution and stabilization (Kaiser and Guggenberger, 2000), and aggregation (Haynes and Swift, 1990; Tisdall and Oades, 1982), which physically protect the OM. 2.4.3. Chemically stabilized C (CSOC) The determination of recalcitrant C such as chemically stabilized C (CSOC) was based on the method by Jagadamma et al. (2010). Briefly, 1 g of bulk soil was wetted with 10 mL of distilled water for 10 min. Then 30 mL of H2O2 at 10% was added, and the solution was kept at 50 °C by using a water bath. Each sample was manually shaken daily to ensure a good oxidation, and additional H2O2 was added if necessary. The oxidation period, using H2O2 as the oxidizing agent, requires several days, and depends on texture, mineralogy, the pre-existing SOC concentration, and the nature and quantity of the C inputs. The oxidation was stopped when the frothing completely subsided. The sample was then washed thrice with 30 mL distilled water, and dried at 40 °C until constant weight. The sample was finely ground for C determination. The CSOC constitutes a stable, refractory, pool largely enriched in alkyl C (Jagadamma et al., 2010). 2.4.4. Particulate organic carbon (POC) and mineral-associated organic carbon (MAOC) The particle size fractionation of SOC was performed according to the method described by Feller (1994) and adapted by Sá et al. (2001). This procedure is based on the separation of POC (i.e., 53 to 2000 μm) and MAOC (i.e., b53 μm) by wet sieving. Briefly, 40 g of oven dry (40 °C) bulk sample sieved through a 2-mm sieve, from each treatment and each depth, was prewetted overnight at 8 °C in 100 mL of distilled H2O with sodium hexametaphosphate. Aggregate disruption was accomplished by horizontal shaking at a frequency of 100 rpm with three shaking agate balls with 10-mm diameter for 4 h for samples from the LRV site and for 6 h for those from the PG site. The dispersed soil suspension was wet-sieved with a 53-μm sieve to obtain the 53 to 2000-μm and b53-μm size fractions. The material remaining on the 53-μm sieve was washed with distilled water to obtain the POC. The suspension in the slurry, representing the silt + clay fraction, was transferred in a 1-L glass cylinder, flocculated with CaCl2, and it represented the MAOC (b 53-μm fraction). Both fractions were dried at 40 °C and finely ground for C determination. The POC represents a labile fraction (Cambardella and Elliott, 1994) largely involved as a main source of nutrients for soil microorganisms, in plant nutrition, and in soil structural stability on short-term basis. Meanwhile, the OM associated with finer fractions (i.e., fine silt and clay size), and with lower turnover rates, is considered as the most stabilized pool (Balesdent et al., 1998). 2.4.5. Total organic carbon (TOC) concentration and stocks Sub-samples of 2-mm bulk soil, of aggregate fractions (POC and MAOC) and of CSOC were finely ground (b150 mm), and were used

217

to determine SOC by the dry combustion method using an elemental CN analyzer (TruSpec CN, LECO, St Joseph, USA). The SOC pools were computed to 1-m depth on an equivalent soil mass–depth basis (Ellert and Bettany, 1995) to correct for differences in ρb among land uses (Sá et al., 2013). The rates (Mg ha−1 year−1) of C change among NV and CT, and among NT and CT, were estimated by using the following equation: Depletion rate ¼ ðSOCNV –SOCCT Þ=t

ð1Þ

Recovery rate ¼ ðSOCNT –SOCCT Þ=t

ð2Þ

where SOCNV is the C stock in the field under NV, SOCNT is the C stock in the field under NT, SOCCT is the C stock in the field under CT and t represents the time (in years) since the conversion from NV to CT, and from CT to NT. In addition, a RI (Dieckow et al., 2009; Herrick and Wander, 1997) was calculated to assess SOC recovery rate of different fractions under NT systems. This index uses NV as the upper limit and CT as the lower limit of SOC levels: RI ¼ ðSOCNT –SOCCT Þ=ðSOCNV –SOCCT Þ

ð3Þ

2.5. Statistical analyses Results of SOC concentration, and SOC stocks for different fractions were submitted to an analysis of variance (ANOVA) to assess the significance of the effects of land use and soil tillage systems at each site. Differences among treatments were considered significant at P ≤ 0.05. Statistical calculations were carried out by using R version 2.11.1 (R Development Core Team, 2006), package aov. SigmaPlot 12.0 was used for graphic representation. 3. Results 3.1. PG site 3.1.1. Soil organic carbon pools (TOC, CSOC, HWEOC, and C in the TPS) The TOC concentration was highly stratified with depth under NV (Fig. 1) with 80.7 g kg−1 and 30.6 g kg−1 in the 0–5 cm and 10–20 cm depths, respectively. Similarly, TOC concentration under NT was stratified within the top 20 cm layer, while a uniform concentration was observed under CT due to the mixing and harmonizing effect of successive plowing. The soil under NT was enriched in SOC and contained 18.4 g kg− 1 more TOC than that under CT in the 0–10 cm layer at the PG site. No significant differences in SOC concentration were observed among tillage treatments in the 10–100 cm layer. However, a decreasing trend in TOC concentrations was observed under NV in the subsoil layers. The amounts of C removed by the H2O2 treatment were significantly (P b 0.05) higher in soil under NV (64.4 g kg−1) and NT (53.5 g kg−1) than that under CT, where the amount of labile C removed by H2O2 was 27.4 g kg−1 at the 0–5 cm depth. A similar CSOC concentration was observed among CT, MT and NT treatments in the 5–10 cm until 60 cm depth, despite contrasting TOC concentration. The % of CSOC in TOC increased with increase in depth, and ranged on average among cropped fields from 16% to 35% in the 0–5 cm and in the 80–100 cm depths, respectively. The proportion of the SOC left after H2O2 treatment ranged under NV from 20% (16.3 g kg− 1) in the 0–5 cm depth to 51% (8.5 g kg− 1) in the 80–100 cm depth. The HWEOC and C in the TPS exhibited the same pattern as TOC in the 0–5 cm depth. HWEOC concentration accounted for a small percentage of the TOC, and varied in the soil profile according to the variation in TOC and CTPS fractions (Fig. 1). This trend was also confirmed by the positive correlation between HWEOC and TOC across all soil layers and for each land use (data not shown). The HWEOC concentration

218

F. Tivet et al. / Geoderma 209–210 (2013) 214–225

Hot water extractable organic C (g kg-1)

Soil organic C (g kg-1) 0

20

40

60

100 0,0

80

0,5

1,0

1,5

2,0

Organic C in TPS (g kg-1)

2,5 0

5

10

15

20

0-5

Soil depth (cm)

5-10

10-20

20-40

TOC_NV TOC_CT TOC_MT TOC_NT CSOC

40-60

60-80

NV CT MT NT

80-100

a

b

c

Fig. 1. Soil organic carbon concentration (g kg−1) to 100 cm depth managed under either no-till (NT), conventional tillage (CT), minimum tillage (MT) and under the neighboring native vegetation (NV), at the PG site: a) total soil organic C (TOC), and chemically-stabilized organic C (CSOC), b) hot-water extractable organic C (HWEOC) and c) organic C in total polysaccharides (C in TPS). Horizontal bars denote significance (P = 0.05) among land use.

decreased significantly with increase in depth independent of land uses and tillage treatments. The concentration of HWEOC and CTPS decreased from 2.2 and 16.0 g kg−1 under NV to 0.9 and 8.8 g kg−1 under CT in the 0–5 cm depth, a decrease of 56% and 45% of HWEOC and CTPS, respectively. The ratio HWEOC:TOC decreased sharply under CT from 0.027 (0–5 cm) to 0.012 (10–20 cm). In contrast, values close to or even higher than those for NV were observed in soil under NT (0–20 cm). As already observed in the case of TOC, the CTPS concentrations were also uniformly distributed in the surface layers (0–20 cm) under CT due to the incorporation of crop residues.

Averaged across soil layers, the SOC depleted during fractionation amounted to 3.94, 2.43, 2.86 and 4.49 g kg−1 under NV, CT, MT and NT, respectively. The silt + clay-associated C on the whole represented on average 90.1% (NF) to 93.8% (CT) of the soil mass and 92.0% (NV) to 96.0% (CT) of the SOC fraction mass. The POC, averaged across all soil depths, accounted for 9.9%, 7.4%, 7.0% and 7.1% of the total soil mass under NV, CT, MT and NT, respectively, and SOC represented 3.22 (8.0%), 1.71 (4.0%), 1.87 (5.8%) and 2.55 g kg−1 (5.9%) under NV, CT, MT and NT, respectively. Independent of land uses and tillage treatments, the proportion of MAOC concentration increased slightly with increase in depth in the profile, and represented 90% of TOC under cropped fields and 86% under NV in the 80–100 cm depth. Both SOC fractions were markedly affected by land use conversion and tillage treatments. The adoption of CT reduced TOC concentration to

3.1.2. Particulate and mineral-associated organic carbon concentrations On average, the sum of SOC concentration in both fractions was lower than that of the SOC in bulk soil by 8.4% (CT) to 12.9% (NT). Particulate organic carbon (g kg-1) 0

0

2

4

6

8

10 12 14 16 18

20

0

0

20

40

60

Particulate organic carbon (g kg ) 0

20

20

Mineral associated organic carbon (g kg-1)

-1

Mineral associated organic carbon (g kg-1) 0

5

10

0

02

15

4

6

20

8

0

20

40

60

40

NV CT MT NT

60

80

40

60

b 100

100

10

15

0

0

20

12

14

25

16

5

40

10

15

80

80

a 100

Soil depth (cm)

Soil depth (cm)

5

0

10

15

60

20

Cerrado CT NT1 NT2 NT3 NT4 NT5 NT6

20

80

c

100

d

Fig. 2. Particulate and mineral-associated organic C (g kg−1) to 100 cm depth under NT, CT, MT and under the neighboring native vegetation (NV) at the PG site (a, b), and under CT, NT systems and Cerrado NV at the LRV site (c, d). Error bars for NV indicate standard errors of the means, the horizontal bars refer to the least significant difference (LSD-Student, P = 0.05) between land use.

F. Tivet et al. / Geoderma 209–210 (2013) 214–225

219

Table 1 Soil organic C stock, on an equivalent mass–depth, of total organic C (TOC), hot-water extractable organic C (HWEOC), C in total polysaccharides (CTPS), chemically-stabilized organic C (CSOC), particulate organic C (POC), and mineral-associated organic C (MAOC) in 0–20, 20–40, 40–100, and 0–100 cm depths at the PG site. Soil depth (cm)

Stock (Mg C ha−1)

Land uses

TOC 0–20

20–40

40–100

0–100

a

NV CTb MT NT NV CT MT NT NV CT MT NT NV CT MT NT

HWEOC

92.0 67.4 70.2 84.4 53.7 52.9 49.4 53.2 112.3 129.2 117.1 128.9 258.0 249.5 236.7 266.6

A Bb Bb Aa ns

3.05 1.16 1.54 2.94 1.10 0.41 0.64 0.84 1.78 0.88 0.98 0.58 5.93 2.44 3.17 4.36

ns

ns ab b a

CTPS A Bb Bb Aa A B B AB A B ns B B A Db CDb Ba

CSOC

21.4 16.6 18.6 20.4 16.1 15.8 14.6 15.3 33.0 37.3 35.3 34.6 70.5 69.6 68.4 70.3

A Cc BCb ABa ns

ns

ns

19.4 14.6 11.2 14.9 15.0 13.7 11.3 13.1 50.0 43.3 34.4 42.3 84.4 71.7 56.8 70.4

POC A BCa Cb Ba A Aa Bb ABab A Ba Cb Ba A Ba Cb Ba

MAOC

10.2 5.5 6.1 8.8 2.3 1.7 1.1 0.7 5.7 2.2 2.4 1.7 18.3 9.3 9.5 11.3

A Cb BCb ABa A ABa BCb Cb A Ba Ba Bb A B ns B B

71.0 55.9 56.8 63.6 45.0 46.3 42.3 45.3 94.7 117.5 105.5 112.4 210.7 219.7 204.6 221.3

A Cb BCb ABa ns

B A ns A A ns

a Comparison between tillage systems CT, MT, NT and native vegetation (NV). Uppercase letters indicate difference among NV and tillage treatments. bComparison between tillage systems CT, MT and NT. Lowercase letters indicate difference between tillage treatments.

45.7 g C kg−1, compared with decline by 82% in POC (−10.5 g C kg−1) and 49% in MAOC (−29.3 g C kg−1) in the 0–5 cm depth (Fig. 2). Most of the differences in SOC concentration in both aggregate fractions between NT and CT were observed in the 0–5, 5–10, 10–20, and 20–40 cm depths. Higher POC concentrations were recorded under CT in the 10–20 and 20–40 cm depths when compared with NT (P b 0.05), resulting from the redistribution of SOC by successive plowing. In contrast, higher MAOC concentrations (P b 0.05) were recorded under NT at the 0–5 and 5–10 cm depths. Similar values of MAOC concentration were observed in the 40–100 cm depth under NT and CT. 3.1.3. SOC stocks of different fractions The CTPS and CSOC accounted for 54% of TOC stock up to the 1-m depth under cropped fields emphasizing the importance of both

fractions (Table 1). The HWEOC in the 0–20 cm depth accounted for only a small proportion of TOC stock, and represented over 3.3% under NV, 3.5% under NT, and only 1.7% and 2.1% under CT and MT, respectively. Difference in SOC among NV and CT was mainly observed in the 0–20 cm layer, while significant differences were observed in labile pools such as HWEOC, and POC to the 1-m depth. Similarly, SOC in NT in the 0–20 cm depth was close to NV, and also significantly higher than that under CT. In addition, HWEOC was significantly higher under NT than that under CT in the 0– 20 cm depth, and when considering the entire profile to the 1-m depth. The POC and MAOC stocks in the 20-cm depth represented 1.9 times (4.74 Mg C ha−1) and 1.3 times (15.10 Mg C ha−1) more C under NV than under CT, respectively. At the end of 29 years, POC (5.5 vs. 8.8 Mg C ha− 1, P b 0.01) and MAOC (55.9 vs. 63.6 Mg C ha− 1,

160

Resilience index (0-20 cm) POC 0.71 MAOC 0.52

P = 0.008

120

A A B

POC MAOC

80

0-20 cm

100 A P = 0.001

A

B

60 A

A

B

P = 0.001

40

20

42 years Pasture (10 yrs) + CT (32 yrs)

20-40 cm

Soil organic C stock (Mg ha -1)

140

P = 0.544

42 years Pasture (10 yrs) + CT (3 yrs) + NT (29 yrs)

0 CT

Annual C input: 3.07 Mg C ha-1

NV (nd)

NT 4.15 Mg C ha-1

Fig. 3. Particulate organic C (POC) and mineral-associated organic C (MAOC) stocks (Mg C ha−1) in 0–20 cm and 20–40 cm depths under native vegetation (NV), conventional tillage (CT), and no-till (NT) systems at the PG site. SOC fractions followed by different uppercase letter are significantly different across land uses based on LSD-Student. The level of significance for each soil depth interval is indicated for both SOC fractions. The SOC resilience index ([SOCNT − SOCCT]/[SOCNV − SOCCT]) is given for both SOC fractions for 0–20 cm depth interval, where SOCNT is the C stock in the field under NT, SOCCT is the C stock in the field under CT, SOCNV is the C stock in the field under NV. Annual C input of tillage treatments from Sá et al. (2013).

220

F. Tivet et al. / Geoderma 209–210 (2013) 214–225

P b 0.001) stocks in the 0–20 cm depth were consistently lower under CT than those under NT (Table 1 and Fig. 3). In contrast, higher POC stocks were recorded under CT in the 20–40 cm depth (Table 1). In NT soil, the RI increased among SOC pools in the order HWEOC N CTPS N POC N MAOC in the 0–20 cm depth (Table 3).

under NV. The concentrations of HWEOC and CTPS in the 0–5 cm depth decreased from 0.8 and 11.3 g C kg−1 under NV to 0.4 and 6.5 g C kg−1 under CT, a reduction of 52% and 42%, respectively. Among tillage treatments, higher concentrations of HWEOC in the 0–5 cm depth were observed under NT (P b 0.05), and higher concentrations of CTPS were observed under all NT systems when compared with CT from the 0–5 to 40–60 cm depths.

3.2. LRV site

3.2.2. Particulate and mineral-associated organic carbon concentrations Averaged across soil layers, the difference between TOC and the sum of both aggregate fractions ranged from 0.20 to 1.20 g C kg−1 of bulk soil. The silt + clay-associated C as a whole represented on average 48.6% (CT) to 51.2% (NV) of the soil mass, and contained 81.7% (Cerrado NV) to 84.9% (NT4) of the SOC fraction mass. In contrast with the sub-tropical PG site, the POC fraction accounted for a higher proportion of total aggregate fractions, accounting for 48.8% (NV) to 51.4% (CT) of the total soil mass, and representing 15.1 to 18.3% of the SOC fraction mass. Conversion from Cerrado NV to CT reduced TOC to 20.0 g C kg−1 in 0–5 cm depth, with estimated loss of 71% for POC (−11.2 g kg−1) and 40% for MAOC (−8.9 g kg−1) (Fig. 3). The MAOC concentration was significantly lower in soil under CT than that under Cerrado NV in the 0–5, 5–10, and 20–40 cm depths, while consistently higher concentrations were observed in the 0–40 cm depth interval under NT1, NT5 and NT6. In contrast, no significant differences were observed among NT systems and CT for the POC fraction.

3.2.1. Soil organic carbon pools (TOC, CSOC, HWEOC, and C in the TPS) No significant differences in TOC concentrations were observed among the NT systems (Fig. 4). However, there was a trend of somewhat higher TOC concentration under NT systems (NT1, NT5, and NT6) characterized by a greater biomass-C input. When considering all land uses, significant differences at P b 0.001 (Fig. 4) were observed in the 0–5 cm depth with greater TOC concentration observed under NV than in cultivated soils. Soil under some NT systems (NT4, NT5 and NT6) and NV exhibited higher TOC concentrations (P b 0.05) in the 20–40 cm depth than that under CT. The CSOC concentrations were almost constant in the surface and sub-soil layers (Fig. 4), but differences among treatments were observed for each depth interval (except the 0–5 cm depth). As observed at the PG site, the proportion of the SOC left after H2O2 treatment increased with increase in depth, and ranged under Cerrado NV from 8.3% (3.2 g kg−1) in the 0–5 cm depth to 47.8% (3.3 g kg−1) in the 80–100 cm depth. Similar to the PG site, HWEOC concentrations was also proportional to the TOC and TPS (Fig. 4), but differed across soil layers between NV and cropped fields (CT and NT cropping systems). Independent of land use change and management practices, the concentrations of HWEOC and CTPS decreased slightly with increase in soil depth, and the proportion of CTPS in TOC increased with depth. The concentrations of HWEOC and CTPS were significantly affected by the conversion from Cerrado NV to cropped fields with a higher concentration of both SOC pools observed

Hot water extractable organic C (g kg-1)

Soil organic C (g kg-1) 0

10

20

30

3.2.3. SOC stocks of different fractions The CTPS and CSOC accounted for 52% of TOC stock under cropped fields (NT and CT), while CSOC pool represented, among land uses, almost 30% of TOC to the 1-m depth (Table 2). Under NV, 78% (102.2 Mg C ha−1) of the total C stock to the1-m depth was stabilized in the MAOC fraction, and ranged from 79% (NT1) to 84% (CT)

50 0,0

40

0,2

0,4

0,6

1,0 0

0,8

Organic C in TPS (g kg-1) 1

2

3

4

5

6

0-5

Soil depth (cm)

5-10

10-20

20-40

TOC_NV TOC_CT TOC_NT3 TOC_NT2 TOC_NT4 TOC_NT6 TOC_NT1 TOC_NT5 CSOC

40-60 NV CT NT3 NT2 NT4 NT6 NT1 NT5

60-80

80-100

a

b

c

Fig. 4. Soil organic carbon concentration (g kg−1) to 100 cm depth under no-till systems (NT1-6), conventional tillage (CT) and under the neighboring native vegetation of Cerrado (NV), at the LRV site: a) total organic C (TOC), and chemically-stabilized organic C (CSOC), b) hot-water extractable organic C (HWEOC) and c) organic C in total polysaccharides (C in TPS). Horizontal bars denote significance (P = 0.05) among land use.

F. Tivet et al. / Geoderma 209–210 (2013) 214–225

221

Table 2 Soil organic C stock, on an equivalent mass–depth, of total organic C (TOC), hot-water extractable organic C (HWEOC), C in total polysaccharides (CTPS), chemically-stabilized organic C (CSOC), particulate organic C (POC), and mineral-associated organic C (MAOC) in 0–20, 20–40, 40–100, and 0–100 cm depths at the LRV site. Soil depth (cm)

Land uses

Stock (Mg C ha−1) TOC

a

0–20

NV CT NT1b NT2 NT3 NT4 NT5 NT6 NV CT NT1 NT2 NT3 NT4 NT5 NT6 NV CT NT1 NT2 NT3 NT4 NT5 NT6 NV CT NT1 NT2 NT3 NT4 NT5 NT6

20–40

40–100

0–100

HWEOC

48.0 33.8 44.2 39.5 37.7 37.7 43.3 40.7 27.7 21.8 25.1 25.8 23.7 27.5 28.1 26.5 54.6 48.5 49.9 49.2 48.8 51.0 49.1 51.4 130.3 104.1 119.2 114.5 110.3 116.1 120.5 118.7

A C AB ns BC BC BC AB AB A C ABC ns ABC BC AB AB AB ns

A C AB ns BC BC C AB AB

1.16 0.37 0.57 0.49 0.52 0.40 0.91 0.64 0.66 0.15 0.23 0.35 0.32 0.23 0.53 0.46 0.85 0.18 0.34 0.56 0.55 0.36 0.92 0.47 2.68 0.69 1.14 1.41 1.40 0.98 2.36 1.56

CTPS A D CDbc CDbc CDbc CDc ABa BCb A C C ns BC BC C AB AB AB C C ns ABC ABC C A B A B Bb Bb Bb Bb Aa Bb

8.1 5.1 7.9 7.6 7.5 7.3 7.4 7.5 7.1 4.9 6.8 7.6 7.2 7.1 6.8 7.0 14.3 11.4 14.3 15.1 13.8 12.5 13.0 13.2 29.4 21.3 29.0 30.3 28.6 26.9 27.2 27.7

CSOC A B A ns A A A A A ns

A B A ns A A AB AB AB A B A ns A A A A A

6.3 5.4 7.1 6.7 5.9 6.9 7.4 6.4 6.5 5.8 6.8 6.8 6.6 7.8 6.9 6.9 22.0 17.1 20.9 22.2 19.6 22.8 22.6 21.0 34.8 28.3 34.8 35.7 32.1 37.4 36.8 34.4

POC BCD D AB ns ABC CD ABC A ABC B C B ns B B A B B A C ABab Aa Bb Aa Aa ABab ABC D ABab Aba CDb Aa Aba Bab

12.7 6.9 9.8 8.1 6.9 6.4 8.8 9.3 3.7 2.5 2.6 2.7 2.2 2.6 3.1 2.6 6.0 6.1 5.9 5.8 5.6 5.7 6.1 6.6 22.4 15.5 18.4 16.6 14.8 14.7 18.0 18.5

MAOC ns

ns

ns

ns

34.0 26.2 30.6 29.4 28.1 27.8 31.8 30.3 22.6 19.1 20.8 22.0 20.8 20.5 21.6 22.6 45.5 42.2 42.2 42.3 40.9 40.7 42.4 43.4 102.2 87.4 93.7 93.7 89.9 89.0 95.9 96.3

A C ABC ns BC BC BC AB ABC A B AB ns A AB AB A A ns

A C B ns BC BC BC AB AB

a Comparison between tillage systems CT, NT systems, and native vegetation (NV). Uppercase letters indicate difference among NV and tillage treatments. b Comparison between NT systems (NT1-6). Lowercase letters indicate difference between NT systems.

under tillage treatments. The proportion of MAOC among land uses and management practices increased in the C-depleted subsoil layers and represented on average 87% of TOC at the 80–100 cm depth. Conversion of NV to CT depleted labile and stable SOC fractions in the 0–20 cm depth, with 68%, 37%, 46%, and 23% decreases on HWEOC, CTPS, POC, and MAOC fractions, respectively. Higher values of each SOC fractions were observed under NT systems when comparing with CT, with the exception of POC pool under NT4. In NT soils, RI increased among SOC pools in the order CTPS N MAOC N HWEOC N POC at the 0–20 cm depth (Table 3), and with higher values observed under NT1, NT5, and NT6. At the end of 8 years, POC (9.5 vs. 11.9 to 12.4 Mg C ha−1, P =0.11) and MAOC (45.3 vs. 51.5 to 53.4 Mg C ha−1, P b 0.05) in the

Table 3 Resilience index of total soil organic carbon (TOC) and fractions (HWEOC, CTPS, POC, and MAOC) under MT and NT at the PG site, and under NT cropping systems at the LRV site in 0–20 cm depth. Site

PG LRV

a

Tillage

MT NT NT1 NT2 NT3 NT4 NT5 NT6

a

Resilience index

b

TOC

HWEOC

0.11 0.69 0.73 0.40 0.27 0.27 0.67 0.49

0.20 0.94 0.26 0.16 0.19 0.04 0.69 0.34

c

CTPS

POC

MAOC

0.41 0.79 0.93 0.82 0.81 0.74 0.78 0.81

0.13 0.71 0.50 0.19 0.00 −0.10 0.32 0.40

0.06 0.52 0.57 0.41 0.25 0.21 0.72 0.53

MT: minimum tillage; NT: no-tillage. b Resilience index = [SOCNT − SOCCT]/ [SOCNV − SOCCT], using the respective SOC stock of each land use and management practices in 0–20 cm depth. c HWEOC: hot-water extractable organic C; CTPS: C in total polysaccharides; POC: particulate organic C; MAOC: mineral-associated organic C.

0–40 cm layer were consistently lower under CT than those under NT1, NT5, and NT6 (Fig. 5). 4. Discussion 4.1. Changes in chemically stabilized organic C, C in total polysaccharides, and hot-water extractable organic C concentrations Despite significant differences observed among land use, the CSOC concentration was almost constant throughout the soil profile, with the exception of the 0–5 cm depth at the PG site. In general, CSOC in the surface layer is comprised of silt and clay-associated C with prevalent recalcitrant macromolecules but also of fresh aliphatic plant materials which are resistant to H2O2 oxidation (Eusterhues et al., 2005). The oxidation, done on bulk soil without removal of the labile C pool prior to the treatment, could explain this trend in the soil surface layer. At both sites, the amount of young plant residue-derived SOC did not affect the concentration of CSOC. Thus, the CSOC in the C baseline could be considered as closely linked to the morphology and chemical structure of OM, the chemical and physical nature of the soil mineral fraction and the architecture of the soil matrix. The CSOC represented on average, across land uses, 7.2 g C kg−1 at the PG site and 3.1 g C kg−1 at the LRV site. In the present study, the proportion of CSOC increased consistently in C-depleted subsoils among land uses and management practices. An increase in proportion of a refractory C pool in the C-depleted subsoil is reported extensively in the literature. For bulk soil under a temperate climate, Jagadamma and Lal (2010) observed that the proportion of TOC pool as a refractory C pool increased from ~5% in surface layers to ~29% in subsoil among contrasting land uses. Soil carbohydrates constitute a significant part of the labile SOC pool indicating a large contribution of non-cellulosic and cellulosic

222

F. Tivet et al. / Geoderma 209–210 (2013) 214–225 80

Resilience index (0-20 cm) POC MAOC

0.00 0.25

-0.10 0.21

0.19 0.41

0.40 0.53

0.50 0.57

0.32 0.72

POC

40

MAOC

C

0-20 cm

60

A

P = 0.04

BC BC

BC

AB

AB

NT3 6.84

NT4 7.34

ABC

AB ABC

P = 0.31 20

B

Conversion + CT23 yrs 0

CT

Annual C input: 4.0 Mg C ha-1

20-40 cm

-1

Soil organic C stock (Mg ha )

P = 0.12

A

P = 0.05

A

A

NT2 7.25

NT6 7.41

AB

A

Conversion + CT 15 yrs + NT8 yrs Cerrado NV 4.3 Mg C ha-1

NT1 7.60

NT5 8.38 Mg C ha-1

Fig. 5. Particulate organic C (POC) and mineral-associated organic C (MAOC) stocks (Mg C ha−1) in 0–20 cm and 20–40 cm depths under native vegetation (Cerrado NV), conventional tillage (CT), and no-till (NT 1–6) systems at the LRV site. MAOC fraction followed by different uppercase letter is significantly different across land uses based on LSD-Student. The level of significance for each soil depth interval is indicated for both SOC fractions. The SOC resilience index ([SOCNT − SOCCT]/[SOCNV − SOCCT]) is given for both SOC fractions for 0–20 cm depth interval, where SOCNT is the C stock in the field under NT, SOCCT is the C stock in the field under CT, SOCNV is the C stock in the field under NV. Annual C input of tillage treatments from Sá et al. (2013) and of Cerrado NV from Corbeels et al. (2006).

polysaccharides in the soil. Soil under NT (0–20 cm) had HWEOC concentration close than those under NV at the PG site, leading to a clear formation of a gradient from surface to sub-soil layers, which is a characteristic of a native vegetation. These trends were less evident at the LRV site, but higher values of HWEOC in the 0–5 cm depth were observed under NT systems when compared with CT. Several studies (i.e., Ball et al., 1996; Beare et al., 1997), showed an enrichment of microbially derived carbohydrates under NT versus CT. A greater proportion of microbial-derived compared with plant-derived carbohydrates under NT compared with CT may be due to the relatively higher fungal biomass under NT (Frey et al., 1999). In addition, incorporation of forage species into crop rotations (NT1, NT5 and NT6) seems to increase HWEOC in the 0–5 cm depth, probably due to higher root inputs and the stimulation of microbial activity that follows (Lienhard et al., 2012). Also, HWEOC is associated with high concentration of soluble carbohydrates, which is more likely to be present in systems with a high annual C input. In contrast, a constant proportion of SOC present as acid hydrolyzable carbohydrates was observed across differing NT systems at the LRV site. Since HWEOC are considered more microbially derived than acid hydrolyzable carbohydrates, the microbially derived carbohydrates seem to be more affected by tillage (at both locations) and by NT cropping sequences than plant-derived carbohydrates. Thus, a large variation in the HWEOC concentration observed among land uses and management corroborates the importance of HWEOC as a soil quality indicator to characterize contrasting land uses and tillage treatments (Ghani et al., 2003). 4.2. Changes in POC and MAOC concentrations In both sites, land uses with higher TOC concentration exhibited greater amount of SOC lost during fractionation, emphasizing greater water soluble C under NV and NT systems. Metay et al. (2007), reporting the concentration of all SOC fractions, concluded that water soluble SOC, obtained through fractionation, ranged from 8% to ~15% of the TOC of bulk soil for a clayey Oxisol of the Cerrado region. Chivenge et al. (2007) reported losses of SOC during fractionation from 1.7 to 3.0 g kg−1 on a red clay soil, representing 8% to 18% of TOC in the bulk soil. The MAOC is largely enriched in polysaccharides and CSOC in subsoil layers. Averaged across land uses, CTPS and CSOC in the 80–100 cm

depth at the PG site accounted approximately for 28% and 41% of MAOC, respectively (Table 1). For the same soil depth interval, CTPS and CSOC represented 14% and 36% of MAOC at the LRV site, respectively (Table 2). Similarly, Wattel-koekkoek et al. (2001) indicated that kaolinite-associated SOC is enriched in polysaccharide products for kaolinitic soils from Brazil, Kenya, Mali and Mozambique. Thus, the presence of easily degradable substances such as polysaccharides in this fraction in subsoil layers emphasized that spatial inaccessibility within microaggregates and physical protection provided by the mineral matrix plays a crucial role in SOC stabilization (Rumpel et al., 2010). The SOC fractions are negatively affected by CT in the soil surface layers of these two clayey soils, principally in the 0–5 cm depth for POC, and 0–5 and 5–10 cm depths for MAOC. This result is consistent with the study conducted by Feller and Beare (1997), which also showed that, under continuously cultivated soils, the decrease in SOC is primarily due to a loss of POC in sandy soils and of clay-associated C in clayey soils. At the LRV site, much of the variation in SOC among CT and NT systems is due to the silt + clay-associated C, indicating that a significant proportion of the SOC is relatively labile, protected within microaggregates and/or through interactions with minerals. 4.3. Comparison among sub-tropical and tropical sites The MAOC (210.7 Mg ha−1) and CSOC (84.4 Mg ha−1) pools in the sub-tropical Rhodic Hapludox under NV were more than twice as high as the respective pools in the Cerrado Typic Haplustox (MAOC = 102.2 Mg ha−1; CSOC = 34.8 Mg ha−1), while the POC pool was comparable but higher at the LRV compared with the PG site: 22.4 vs. 18.3 Mg ha−1. Thus, relatively more POC could be mineralized in the tropical Typic Haplustox due to the higher proportion of this fraction. The high clay content, and Al and Fe-sesquioxides at the PG site (Goncalves et al., 2008) increase the specific surface area, micro pores and the potential stabilization of OM against biological processes, and are the principal determinants of a higher MAOC and CSOC pools under this sub-tropical climate. In addition, several studies, including that by Shang and Tiessen (1998) and Barthes et al. (2008), have highlighted the role of both crystalline and non-crystalline Fe and Al-sesquioxides in stabilizing the SOC pool through the formation

F. Tivet et al. / Geoderma 209–210 (2013) 214–225

223

Fig. 6. Relationships between the annual C input (above and belowground) and the resilience index (0–20 cm) for TOC (a), POC and MAOC (b) at the PG and LRV sites.

of organo-mineral complexes. It was also observed that the fluorescence spectra of the sub-tropical Rhodic Hapludox (PG site) had a fragmented shape probably due to more complex interactions between organic compounds and minerals, when compared with soils at the LRV site (Tivet et al., 2013b). Thus, recalcitrance of organic compounds is higher in soils at the LRV site when organo-mineral interactions are predominant on selective preservation at the PG site. The results presented indicate that CSOC constitutes a significant part of the SOC stock at both sites, representing on average 28% to 29% to the 1-m depth at the PG and LRV sites, respectively. For a clayey Oxisol in the Cerrado region, Jantalia et al. (2007) also observed a refractory C pool under NV and tillage treatments of 65.5 and 49.1 Mg ha−1, representing 32% and 29% of the SOC, respectively. Jagadamma and Lal (2010) conducted an analysis to assess the relevance of chemical and physical methods of isolating the stable SOC pool. For a soil in the temperate climate, chemical oxidation of bulk soil with H2O2 and physical fractionation plus chemical oxidation with H2O2 were efficient and similar in isolating the stable SOC fractions in the 30– 45 cm layer. These results support the assumption that 100% of the CSOC was included in the MAOC fraction in the 60–100 cm depth, indicating that part of the MAOC is susceptible to chemical oxidation by H2O2. The proportion of the non-chemically stabilized C from MAOC decreases with increase in soil depths, and shows a similar trend among sub-tropical and tropical regions. Under NV, the non-chemically stabilized C accounted for 54% (9.8 g kg− 1) and 57% (4.3 g kg− 1) of the MAOC in 40–60 cm depth at the PG and LRV sites, and decreased to 41% (6.0 g kg− 1) and 47% (2.9 g kg− 1) in 80–100 cm depth, respectively. The difference in the SOC stocks between CT and NV in the 0- to 40-cm layer (Tables 1 and 2) represented an average rate of loss of 0.60 Mg C ha−1 yr−1 for the clayey Rhodic Hapludox (PG, 42 years under CT, ~ 697 g clay kg−1) and 0.88 Mg C ha−1 yr−1 for the clayey Typic Haplustox (LRV, 23 years under CT, ~403 g clay kg−1). It is widely reported that the texture and mineral composition played a key role in the decrease of the original SOC stock after conversion to CT, so that as the clay content decreased, the rate of SOC loss increased (Dalal and Mayer, 1986). Results obtained in the present study show that changes induced by CT in the 0–20 cm depth decreased SOC stocks with different turn-over time. The HWEOC and POC stocks for the entire 1-m soil profile presented, respectively, 59% (3.49 Mg C ha− 1) and

49% (8.92 Mg C ha− 1) less SOC under CT than under NV. Despite the high stable microaggregates, due to high clay content, Fe and Al-sesquioxides, MAOC was also significantly affected by CT at the PG site, reflecting the low sustainability of the CT system. The rate of loss of SOC fractions in the 0–20 cm depth at the LRV site over the 23 years of CT was 0.25 and 0.34 Mg C ha−1 yr−1 for POC (46%) and MAOC (23%), respectively. However, despite the negative effect of conventional plow-based tillage, an important effect to be emphasized is the possibility of recovering most of the SOC fractions by adopting high biomass-C inputs under NT management, and despite the fact that the experimental duration at the LRV site was only eight years. When considering NT with high biomass input (i.e., NT1, NT5 and NT6), the gain in SOC ranged from 0.23 to 0.36 Mg C ha−1 yr−1 for POC and 0.52 and 0.70 Mg C ha−1 yr−1 for MAOC. Both SOC fractions were restored faster than the rate of soil degradation which occurred after conversion of NV. At both locations, the values of the aggregation parameters (i.e. mean weight diameter, aggregate stability index) were close under NT soils than those under NV at the soil surface layers, indicating the potential of NT to restore aggregation (Tivet et al., 2013a). In addition, the results of humification index, through Laser-Induced Fluorescence spectroscopy (LIFS), at the PG and LRV sites suggest that the SOC associated with macro and microaggregates in soils under NV and NT may be qualitatively more labile and less processed than that associated with the macro and microaggregates under CT (Tivet et al., 2013b). We also observed that the adoption of NT cropping systems changed significantly the composition of organic-C compounds at both locations, with greater values of aliphatic-C observed under NT than that under CT in soil surface layers (Tivet et al., 2013b). This is a clear indication that selective preservation and organo-mineral interactions are not the major mechanisms promoting C accumulation in NT soils, and that spatial inaccessibility within aggregates played a key role in SOC accumulation (Tivet et al., 2013a) at both locations. The difference in SOC pool among CT and NT systems is largely caused by soil C storage in physically protected fraction in the 0–20 cm and 20–40 cm depths. Sá et al. (2001) also observed that a significant contribution of crop residues to SOC in the 0–10 cm depth occurred in the silt-sized and finer clay fractions under a 10-year NT system, indicating that most of the young SOC was protected in the MAOC fraction. A higher RI was observed for CTPS (at both sites) and HWEOC (mainly at the PG site), which are the main binding agents for

224

F. Tivet et al. / Geoderma 209–210 (2013) 214–225

aggregation, and which may also reinforce the SOC storage by reducing the rate of mineralization. SOC accumulation (0.60 Mg C ha−1 yr−1 in the 0–40 cm depth) under NT and resilience index for MAOC (Table 3) is relatively low at the PG site, when compared with the rate of C sequestration in the region that ranges from 0.66 Mg C ha−1 yr−1 (Pavei, 2005; 0–20 cm depth), 0.8 Mg C ha−1 yr−1 (Neto et al., 2009; 0–40 cm depth), and ~1.0 Mg C ha−1 yr−1 (Sá et al., 2001; 0–40 cm depth; Venzke-Filho et al., 2004; 0–20 cm depth). In contrast, the biodiversity and biomass-C inputs of the NT systems at the LRV site may support a continuous and higher flow of mass and energy, which releases organic compounds, accentuates soil biodiversity, and enhances on a short-term period SOC recovery (Séguy et al., 2006; Uphoff et al., 2006). The rate of SOC sequestration in the 0–40 cm depth ranged from 0.73 to 1.98 Mg C ha−1 yr−1 at the LRV site (Sá et al., 2013). In addition, the RI for MAOC increased significantly with the increase in annual C input (Fig. 6). Additional research is needed to analyze the cumulative effects of a mixture of cover/relay crops under NT on root exudation, and changes in rhizospheric biological activities, which influence nutrients and C cycling by improving soil aggregation and physically protecting plant-derived SOC and microbially derived organic C. In an acid tropical grassland environment in Laos, Lienhard et al. (2013) showed that high and diversified biomass-C inputs under NT induced significant modifications when compared with conventional plow-based tillage. NT increased aggregate stability and soil organic carbon content, enhanced nutrient availability and microbial biomass as a result of a simultaneous increase of fungal and bacterial densities. They also showed a discrimination of soil microbial community structures between NT, CT, and native grassland vegetation. Several studies have also observed that microbial communities with a predominance of fungal biomass accumulate more C than bacterial-dominated communities (Six et al., 2006), due to a higher recalcitrance of fungal derived by-products (Martin and Haider, 1979), greater transfer of root-derived C to SOC (Godbold et al., 2006), and to stronger interactions with clay minerals and aggregates (Simpson et al., 2004) reducing the decomposition rate. 5. Conclusions With reference to the principal hypothesis, the data presented show that several SOC fractions were negatively impacted by the conversion of NV to CT, and that losses of SOC fractions were restored by the adoption of NT systems. However, the magnitude of recovery in SOC fractions depends of the input of biomass. The results presented support the following conclusions: • Deforestation and conversion to cultivated field under CT reduced both labile (HWEOC, TPS, OXC, and POC) and stable (MAOC, b 53 μm) pools. In contrast, only small changes were observed in the refractory (CSOC) pool which is highly stable and is less impacted by changes in land use and management. • At the PG site, the HWEOC and CTPS concentrations in the 0–5 cm depth decreased by 56% and 45% in CT soil, respectively. The adoption of CT reduced POC by 46% (4.7 Mg ha−1), and MAOC by 21% (15.1 Mg C ha−1) in the 0–20 cm depth. Adoption of NT increased POC by 0.12 Mg C ha−1 yr−1, and MAOC by 0.27 Mg C ha−1 yr−1. • At the LRV site, the concentration of HWEOC and CTPS in the 0–5 cm depth decreased by 50% and 42%, respectively. Following the 23 years of CT at the LRV site, the rate of loss of SOC fractions in the 0–20 cm depth was 0.25 and 0.34 Mg C ha−1 yr−1 for POC and MAOC, respectively. In contrast, the adoption of intensive NT systems increased POC by 0.23 to 0.36 Mg C ha− 1 yr− 1, and MAOC by 0.52 and 0.70 Mg C ha− 1 yr− 1. • The concentration of CSOC was almost constant along the soil profile at both locations (~ 7.2 g kg− 1 at the PG site, and ~ 3.1 g kg− 1 at the LRV site), and among all land use and management. In the

SOC-depleted subsoil, the proportion of CSOC and CTPS increased and represented a large portion of the MAOC. • Intensive NT systems (e.g. diversity of cover/relay crops and high annual biomass input) at the tropical LRV, increased HWEOC, C in the TPS, and enhanced the short-term process of SOC stabilization in the mineral-associated fraction. In addition, relatively higher resilience indices of POC and MAOC fractions were observed under NT systems with high biomass input at the LRV site. • The results presented emphasize the high potential of NT systems to restore SOC fractions previously depleted by CT, contributing to a high farm productivity and sustainability.

Acknowledgments We would like to thank the Agricultural Research Institute of Paraná, Lucas do Rio Verde Foundation, and particularly Jadir Rosa and Clayton Bortolini for allowing the access to the experimental fields. We greatly appreciate the help from Ms. Jaqueline Aparecida Gonçalves and Mr. Romeu Martins Filho for laboratory analyses. This work was supported by the Agrisus Foundation (PA 677-10), the Centre de Coopération Internationale en Recherche Agronomique pour le Développement (CIRAD), and by the Climate, Water, and Carbon project of the Ohio State University. References Balesdent, J., Besnard, E., Arrouays, D., Chenu, C., 1998. The dynamics of carbon in particle-size fractions of soil in a forest-cultivation sequence. Plant and Soil 201 (1), 49–57. Ball, B.C., Cheshire, M.V., Robertson, E.A.G., Hunter, E.A., 1996. Carbohydrate composition in relation to structural stability, compactibility and plasticity of two soils in a longterm experiment. Soil and Tillage Research 39 (3–4), 143–160. Barthes, B.G., Kouakoua, E., Larre-Larrouy, M.C., Razafimbelo, T.M., de Luca, E.F., Azontonde, A., Neves, C., de Freitas, P.L., Feller, C.L., 2008. Texture and sesquioxide effects on water-stable aggregates and organic matter in some tropical soils. Geoderma 143 (1–2), 14–25. Batlle-Bayer, L., Batjes, N.H., Bindraban, P.S., 2010. Changes in organic carbon stocks upon land use conversion in the Brazilian Cerrado: a review. Agriculture, Ecosystems and Environment 137 (1–2), 47–58. Bayer, C., Martin-Neto, L., Mielniczuk, J., Ceretta, C.A., 2000. Effect of no-till cropping systems on soil organic matter in a sandy clay loam Acrisol from Southern Brazil monitored by electron spin resonance and nuclear magnetic resonance. Soil and Tillage Research 53 (2), 95–104. Bayer, C., Martin-Neto, L., Mielniczuk, J., Pavinato, A., 2004. Carbon storage in labile fractions of soil organic matter in a tropical no-tillage Oxisol. Pesquisa Agropecuária Brasileira 39 (7), 677–683. Beare, M.H., Cabrera, M.L., Hendrix, P.F., Coleman, D.C., 1994a. Aggregate-protected and unprotected organic-matter pools in conventional-tillage and no-tillage soils. Soil Science Society of America Journal 58 (3), 787–795. Beare, M.H., Hendrix, P.F., Coleman, D.C., 1994b. Water-stable aggregates and organicmatter fractions in conventional-tillage and no-tillage soils. Soil Science Society of America Journal 58 (3), 777–786. Beare, M.H., Reddy, M.V., Tian, G., Srivastava, S.C., 1997. Agricultural intensification, soil biodiversity and agroecosystem function in the tropics: the role of decomposer biota. Applied Soil Ecology 6 (1), 87–108. Blake, G.R., Hartge, K.H., 1986. Bulk Density, In: Klute, A. (Ed.), Methods of Soil Analysis, Part I, Second ed. Physical and Mineralogical Methods: Agronomy Monograph no. 9. ASA SSSA, Madison, WI, USA, pp. 363–375. Cambardella, C.A., Elliott, E.T., 1994. Carbon and nitrogen dynamics of soil organicmatter fractions from cultivated grassland soils. Soil Science Society of America Journal 58 (1), 123–130. Chivenge, P.P., Murwira, H.K., Giller, K.E., Mapfumo, P., Six, J., 2007. Long-term impact of reduced tillage and residue management on soil carbon stabilization: implications for conservation agriculture on contrasting soils. Soil and Tillage Research 94 (2), 328–337. Corbeels, M., Scopel, E., Cardoso, A., Bernoux, M., Douzet, J.M., Neto, M.S., 2006. Soil carbon storage potential of direct seeding mulch-based cropping systems in the Cerrados of Brazil. Global Change Biology 12 (9), 1773–1787. Dalal, R.C., Mayer, R.J., 1986. Long-term trends in fertility of soils under continuous cultivation and cereal cropping in Southern Queensland. 1. Overall changes in soil properties and trends in winter cereal yields. Australian Journal of Soil Research 24, 265–279. Dieckow, J., Bayer, C., Conceicao, P.C., Zanatta, J.A., Martin-Neto, L., Milori, D.B.M., Salton, J.C., Macedo, M.M., Mielniczuk, J., Hernani, L.C., 2009. Land use, tillage, texture and organic matter stock and composition in tropical and subtropical Brazilian soils. European Journal of Soil Science 60 (2), 240–249.

F. Tivet et al. / Geoderma 209–210 (2013) 214–225 Ellert, B.H., Bettany, J.R., 1995. Calculation of organic matter and nutrients stored in soils under contrasting management regimes. Canadian Journal of Soil Science 75 (4), 529–538. EMBRAPA, 1999. Sistema brasileiro de classificação de solos. Centro Nacional de Pesquisa de Solos. Brasília: Embrapa Produção de Informação.Embrapa Solos, Rio de Janeiro. Eusterhues, K., Rumpel, C., Kogel-Knabner, I., 2005. Stabilization of soil organic matter isolated via oxidative degradation. Organic Geochemistry 36 (11), 1567–1575. Febrapdp, 2011. Federação Brasileira de Plantio Direto na Palha. In Portuguese. Available at: www.febrapdp.org.br. Feller, C., 1994. La matière organique dans les sols tropicaux. University of Strasbourg (ULP), Strasbourg, France. Feller, C., Beare, M.H., 1997. Physical control of soil organic matter dynamics in the tropics. Geoderma 79 (1–4), 69–116. Feller, C., Bernoux, M., 2008. Historical advances in the study of global terrestrial soil organic carbon sequestration. Waste Management 28 (4), 734–740. Frey, S.D., Elliott, E.T., Paustian, K., 1999. Bacterial and fungal abundance and biomass in conventional and no-tillage agroecosystems along two climatic gradients. Soil Biology and Biochemistry 31 (4), 573–585. Ghani, A., Dexter, M., Perrott, K.W., 2003. Hot-water extractable carbon in soils: a sensitive measurement for determining impacts of fertilisation, grazing and cultivation. Soil Biology and Biochemistry 35 (9), 1231–1243. Ghani, A., Muller, K., Dodd, M., Mackay, A., 2010. Dissolved organic matter leaching in some contrasting New Zealand pasture soils. European Journal of Soil Science 61 (4), 525–538. Godbold, D.L., Hoosbeek, M.R., Lukac, M., Cotrufo, M.F., Janssens, I.A., Ceulemans, R., Polle, A., Velthorst, E.J., Scarascia-Mugnozza, G., De Angelis, P., Miglietta, F., Peressotti, A., 2006. Mycorrhizal hyphal turnover as a dominant process for carbon input into soil organic matter. Plant and Soil 281 (1–2), 15–24. Goncalves, D., Leite, W.C., Brinatti, A.M., Saab, S.D., Iarosz, K.C., Masearenhas, Y.P., Carneiro, P.I.B., Rosa, J.A., 2008. Mineralogy of a red Latosol under different management systems for twenty-four years. Revista Brasileira de Ciência do Solo 32, 2647–2652. Haynes, R.J., Francis, G.S., 1993. Changes in microbial biomass-C, soil carbohydratecomposition and aggregate stability induced by growth of selected crop and forage species under field conditions. Journal of Soil Science 44 (4), 665–675. Haynes, R.J., Swift, R.S., 1990. Stability of soil aggregates in relation to organicconstituents and soil–water content. Journal of Soil Science 41 (1), 73–83. Herrick, J.E., Wander, M.M., 1997. Relationships between soil organic carbon and soil quality in cropped and rangeland soils: the importance of distribution, composition, and soil biological activity. CRC Press, Boca Raton, FL. Jagadamma, S., Lal, R., 2010. Distribution of organic carbon in physical fractions of soils as affected by agricultural management. Biology and Fertility of Soils 46 (6), 543–554. Jagadamma, S., Lal, R., Ussiri, D.A.N., Trumbore, S.E., Mestelan, S., 2010. Evaluation of structural chemistry and isotopic signatures of refractory soil organic carbon fraction isolated by wet oxidation methods. Biogeochemistry 98 (1–3), 29–44. Jantalia, C.P., Resck, D.V.S., Alves, B.J.R., Zotarelli, L., Urquiaga, S., Boddey, R.M., 2007. Tillage effect on C stocks of a clayey Oxisol under a soybean-based crop rotation in the Brazilian Cerrado region. Soil and Tillage Research 95, 97–109. Jenkinson, D.S., Ayanaba, A., 1977. Decomposition of C-14-labelled plant material under tropical conditions. Soil Science Society of America Journal 41 (5), 912–915. Kaiser, K., Guggenberger, G., 2000. The role of DOM sorption to mineral surfaces in the preservation of organic matter in soils. Organic Geochemistry 31 (7–8), 711–725. Lal, R., 1976. Soil erosion on alfisols in western Nigeria. 1. Effects of slope, crop-rotation and residue management. Geoderma 16 (5), 363–375. Lal, R., 2004. Soil carbon sequestration impacts on global climate change and food security. Science 304 (5677), 1623–1627. Lienhard, P., Tivet, F., Chabanne, A., Dequiedt, S., Lelievre, M., Sayphoummie, S., Leudphanane, B., Prevost-Boure, N.C., Séguy, L., Maron, P.A., Ranjard, L., 2013. Notill and cover crops shift soil microbial abundance and diversity in Laos tropical grasslands. Agronomy for Sustainable Development 33 (2), 375–384. Lowe, L.E., 1993. Total and labile acid extractable polysaccharide analysis of soils. In: Carter, M.R. (Ed.), Soil Sampling and Methods of Analysis. Lewis, Boca Raton, pp. 373–376. Martin, J.P., Haider, K., 1979. Biodegradation of C-14 labeled model and cornstalk lignins, phenols, model phenolase humic polymers, and fungal melanins as influenced by a readily available carbon source and soil. Applied and Environmental Microbiology 38 (2), 283–289. Metay, A., Moreira, J.A.A., Bernoux, M., Boyer, T., Douzet, J.M., Feigl, B., Feller, C., Maraux, F., Oliver, R., Scopel, E., 2007. Storage and forms of organic carbon in a no-tillage under cover crops system on clayey Oxisol in dryland rice production (Cerrados, Brazil). Soil and Tillage Research 94 (1), 122–132. Neto, M.S., Venzke Filho, S.d.P., Piccolo, M.d.C., Pellegrino Cerri, C.E., Cerri, C.C., 2009. Crop rotation under no-tillage in Tibagi (Paraná State, Brazil). I — Soil carbon sequestration. Revista Brasileira de Ciência do Solo 33, 1013–1022. Paustian, K., Parton, W.J., Persson, J., 1992. Modeling soil organic-matter in organicamended and nitrogen-fertilized long-term plots. Soil Science Society of America Journal 56 (2), 476–488. Pavei, M.A., 2005. Decomposição de resíduos culturais e emissão de gases do efeito estufa em sistemas de manejo do solo em Ponta Grossa (PR). Piracicaba, Universidade de São Paulo, Brazil.

225

R Development Core Team, 2006. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, ISBN 3-900051-07-0. Accessed at www.R-project.org. Rumpel, C., Eusterhues, K., Koegel-Knabner, I., 2010. Non-cellulosic neutral sugar contribution to mineral associated organic matter in top- and subsoil horizons of two acid forest soils. Soil Biology and Biochemistry 42 (2), 379–382. Sá, J.C.D., Cerri, C.C., Dick, W.A., Lal, R., Venske, S.P., Piccolo, M.C., Feigl, B.E., 2001. Organic matter dynamics and carbon sequestration rates for a tillage chronosequence in a Brazilian Oxisol. Soil Science Society of America Journal 65 (5), 1486–1499. Sá, J.C.D., Séguy, L., Tivet, F., Lal, R., Bouzinac, S., Borszowskei, P.R., Briedis, C., Santos, J.B.D., Bertoloni, C., Hartman, D.D.C., Rosa, J., Friedrich, T., 2013. Carbon depletion by plowing and its restoration by no-till cropping systems in Oxisols of sub-tropical and tropical agro-ecoregions in Brazil. Land Degradation and Development. http:// dx.doi.org/10.1002/ldr.2118. Séguy, L., Bouzinac, S., Trentini, A., Côrtes, N., 1998. Brazilian frontier agriculture. Agriculture et Développement. (Special Issue). Séguy, L., Bouzinac, S., Husson, O., 2006. Direct-seeded tropical soil systems with permanent soil cover: Learning from Brazilian experience. In: Uphoff, N., Ball, A.S., Fernandes, E., Herren, H., Husson, O., Laing, M., Palm, C., Pretty, J., Sanchez, P. (Eds.), Biological Approach to Sustainable Soil Systems. CRC Press, Taylor & Francis, pp. 323–342. Shang, C., Tiessen, H., 1998. Organic matter stabilization in two semiarid tropical soils: size, density, and magnetic separations. Soil Science Society of America Journal 62 (5), 1247–1257. Simpson, R.T., Frey, S.D., Six, J., Thiet, R.K., 2004. Preferential accumulation of microbial carbon in aggregate structures of no-tillage soils. Soil Science Society of America Journal 68 (4), 1249–1255. Six, J., Elliott, E.T., Paustian, K., 2000. Soil macroaggregate turnover and microaggregate formation: a mechanism for C sequestration under no-tillage agriculture. Soil Biology and Biochemistry 32 (14), 2099–2103. Six, J., Feller, C., Denef, K., Ogle, S.M., Sá, J.C.D., Albrecht, A., 2002. Soil organic matter, biota and aggregation in temperate and tropical soils — effects of no-tillage. Agronomie 22 (7–8), 755–775. Six, J., Frey, S.D., Thiet, R.K., Batten, K.M., 2006. Bacterial and fungal contributions to carbon sequestration in agroecosystems. Soil Science Society of America Journal 70 (2), 555–569. Soil Survey Staff, 1994. Keys to Soil Taxonomy, Six ed. USDA Natural Resources Conservation Service, US Govt. Printing Office, Washington. Sollins, P., Homann, P., Caldwell, B.A., 1996. Stabilization and destabilization of soil organic matter: mechanisms and controls. Geoderma 74 (1–2), 65–105. Sparling, G., Vojvodic-Vukovic, M., Schipper, L.A., 1998. Hot-water-soluble C as a simple measure of labile soil organic matter: the relationship with microbial biomass C. Soil Biology and Biochemistry 30 (10–11), 1469–1472. Tisdall, J.M., Oades, J.M., 1982. Organic-matter and water-stable aggregates in soils. Journal of Soil Science 33 (2), 141–163. Tivet, F., de Moraes Sá, J.C., Lal, R., Briedis, C., Borszowskei, P.R., dos Santos, J.B., Farias, A., Eurich, G., Hartman, D.d.C., Nadolny Junior, M., Bouzinac, S., Séguy, L., 2013a. Aggregate C depletion by plowing and its restoration by diverse biomass-C inputs under no-till in sub-tropical and tropical regions of Brazil. Soil and Tillage Research 126, 203–218. Tivet, F., Sá, J.C.M., Lal, R., Milori, D.B.M., Briedis, C., Letourmy, P., Pinheiro, L.A., Borszowskei, P.R., Hartman, D.D.C., 2013b. Assessing humification and organic C compounds by laser-induced fluorescence and FTIR spectroscopies under conventional and no-till management in Brazilian Oxisols. Geoderma. http://dx.doi.org/ 10.1016/j.geoderma.2013.05.001. Uphoff, N., Ball, A.S., Fernandes, E., Herren, H., Husson, O., Palm, C., Pretty, J., Sanginga, N., Thies, J., 2006. Understanding the functioning and management of soil systems. In: Uphoff, N., Ball, A.S., Fernandes, E., Herren, H., Husson, O., Laing, M., Palm, C., Pretty, J., Sanchez, P., Sanginga, N., Thies, J. (Eds.), Biological Approaches to Sustainable Soil Systems. CRC Press, Taylor and Francis, pp. 3–13. Venzke-Filho, S.D., Feigl, B.J., Piccolo, M.D., Fante, L., Neto, M.S., Cerri, C.C., 2004. Root systems and soil microbial biomass under no-tillage system. Scientia Agricola 61, 529–537. von Lutzow, M., Kogel-Knabner, I., Ekschmitt, K., Matzner, E., Guggenberger, G., Marschner, B., Flessa, H., 2006. Stabilization of organic matter in temperate soils: mechanisms and their relevance under different soil conditions — a review. European Journal of Soil Science 57 (4), 426–445. von Lutzow, M., Kogel-Knabner, I., Ludwig, B., Matzner, E., Flessa, H., Ekschmitt, K., Guggenberger, G., Marschner, B., Kalbitz, K., 2008. Stabilization mechanisms of organic matter in four temperate soils: development and application of a conceptual model. Journal of Plant Nutrition and Soil Science-Zeitschrift Fur Pflanzenernahrung Und Bodenkunde 171 (1), 111–124. Wattel-Koekkoek, E.J.W., van Genuchten, P.P.L., Buurman, P., van Lagen, B., 2001. Amount and composition of clay-associated soil organic matter in a range of kaolinitic and smectitic soils. Geoderma 99 (1–2), 27–49. Zotarelli, L., Alves, B.J.R., Urquiaga, S., Torres, E., dos Santos, H.P., Paustian, K., Boddey, R.M., Six, J., 2005. Impact of tillage and crop rotation on aggregate-associated carbon in two oxisols. Soil Science Society of America Journal 69 (2), 482–491.