land degradation & developmentLand Degrad. Develop. (2013)
Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/ldr.2218
CARBON DEPLETION BY PLOWING AND ITS RESTORATION BY NO-TILLCROPPING SYSTEMS IN OXISOLS OF SUBTROPICAL AND TROPICAL
AGRO-ECOREGIONS IN BRAZIL
João Carlos de Moraes Sá1*, Lucien Séguy2, Florent Tivet2, Rattan Lal3, Serge Bouzinac2, Paulo Rogério Borszowskei4,Clever Briedis4, Josiane Burkner dos Santos4, Daiani da Cruz Hartman5, Clayton Giani Bertoloni6, Jadir Rosa7,
Theodor Friedrich8
1Department of Soil Science and Agricultural Engineering, State University of Ponta Grossa, Av. Carlos Cavalcanti 4748, Campus de Uvaranas, 84030-900,Ponta Grossa PR, Brazil
2UPR SIA, Centre de Coopération Internationale en Recherche Agronomique pour le Développement, CIRAD, F-34398 Montpellier, France3Carbon Management and Sequestration Center, School of Environment and Natural Resources, The Ohio State University, OARDC/FAES, 2021 Coffey Road,
Columbus, OH 43210, USA4Agronomy Graduate Program, State University of Ponta Grossa, Av. Carlos Cavalcanti 4748, Campus de Uvaranas, 84030-900, Ponta Grossa PR, Brazil5Agronomy Undergraduate Program, State University of Ponta Grossa, Av. Carlos Cavalcanti 4748, Campus de Uvaranas, 84030-900, Ponta Grossa PR,
Brazil6Foundation of Rio Verde, Rodovia MT 449, 78455-000 Lucas do Rio Verde, MT Brazil
7Agricultural Research Institute of Paraná (IAPAR), BR 376, Km 496, 84001-970, Ponta Grossa PR, Brazil8Food and Agriculture Organization, Plant Production and Protection Division (AGP), Room C-782 Viale delle Terme di Caracalla, 00153 Rome, Italy
Received 6 March 2013; Revised 14 March 2013; Accepted 16 March 2013
Agricultural practices can render soil a sink or a source ofatmospheric carbon dioxide, thereby directly influencing thegreenhouse effect (Lal, 2004). The concern about globalwarming (Broecker, 1975) has motivated the scientific com-munity to identify efficient soil management and croppingsystems that can sequester atmospheric CO2 into soil organiccarbon (SOC) (Lal, 2008). Several reports have demonstratedC sequestration in soils managed by a no-till (NT) system inconjunction with complex crop rotation (Rasmussen et al.,1980; Dick, 1983; Kern & Johnson, 1993; Bayer et al.,2000; Sá et al., 2001; Six et al., 2002a; West & Post, 2002;Sisti et al., 2004; Dieckow et al., 2005; Bayer et al., 2006a;Bernoux et al., 2006; Cerri et al., 2007; Sá & Lal, 2009;Boddey et al., 2010). Important factors in increasing CO2
*Correspondence to: J. C. de Moraes Sá, Department of Soil Science andAgricultural Engineering, State University of Ponta Grossa, Av. CarlosCavalcanti 4748, Campus de Uvaranas, 84030-900, Ponta Grossa, PR, Brazil.E-mail: [email protected]
mitigation and the SOC stock are the amount, quality, andfrequency of the crop residues added to soil under a widerange of climate-driven decomposition rates, soil mineralogy,and profile characteristics (Paustian et al., 1997; Six et al.,2002b; Kong et al., 2005; Bayer et al., 2006a; West & Six,2007; Ogle et al., 2012; Virto et al., 2012).The data on SOC sequestration rates for tropical
(�0�03–1�9Mg ha�1 y�1; Bayer et al., 2006a; Corbeelset al., 2006; Cerri et al., 2007; Neto et al., 2010) and subtrop-ical (�0�07–1�4Mgha�1 y�1; (Bayer et al., 2000; Sá et al.,2001; Bayer et al., 2006a; Amado et al., 2006; Cerri et al.,2007; Boddey et al., 2010) regions of Brazil vary widelybecause of the differences in the amount of biomass-C inputand the agronomic potential of agro-ecosystems (Bayer et al.,2006a; Corbeels et al., 2006; Cerri et al., 2007). Batlle-Bayeret al. (2010) reported that changes in the SOC stock, whenconverting from the plow-based conventional tillage (CT)to NT systems, ranged from 0�13 to 1�91Mgha�1 y�1 for a0- to 30-cm depth. A large variability in the SOC sequestrationrate may be attributed to a high diversity of cropping systems,
amount and frequency of biomass-C input, and soil properties(Batlle-Bayer et al., 2010). For example, the rate of SOCsequestration is negligible in those NT systems with lowbiomass-C input and only one crop (summer) per year (Bayeret al., 2006b; Jantalia et al., 2007; Carvalho et al., 2009). Incomparison, in subtropical and tropical ecosystems in Brazil,the rate is high for those NT systems that involve a highbiomass-C input along with a summer crop followed by oneor a combination of several winter crops (Séguy et al., 2006).Furthermore, the rate of SOC sequestration by NT can behigher also in the subsoil layers (Boddey et al., 2010). For adouble cropping system (legume–cereal cropping sequence)under an extensive farmland network, Neto et al. (2010)measured the SOC sequestration rate of 1�98Mgha�1 y�1 forthe 0- to 30-cm depth.The data from some soils of the temperate regions (Baker
et al., 2007; Blanco-Canqui & Lal, 2008; Poirier et al.,2009) show that the increase in SOC stock upon conversionto NT may be limited only to the surface layers along withsome increase in subsoil under CT systems. However, Varvel& Wilhelm (2011) reported an increase in SOC sequestrationeven in subsoil layers by NT cropping in a temperate region.Cropping systems with a high biomass input to maintain a
permanent soil cover mimic the environment under undisturbedecosystems (e.g., forest, savanna, and prairies). These croppingsystems support a continuous flow of mass and energy, whichrelease organic compounds, accentuate soil biodiversity, andenhance soil organic matter (SOM) (Séguy et al., 2006; Sixet al., 2006; Uphoff et al., 2006). These processes are drivenby the multifunctionality of each species in the croppingsystem and its interaction with soil attributes stimulating asystemic interdependence of the soil structure and SOM pools(Séguy et al., 2006; Uphoff et al., 2006). The continuous inputof large amounts of biomass to the soil surface creates a posi-tive C budget, enhances the stable C fraction, and accentuatesC and N transformations and flow (Bayer et al., 2001; Sáet al., 2001; Six et al., 2002a; Denef et al., 2004; Sá & Lal,2009; Boddey et al., 2010). Therefore, the present study wasconducted to test the hypothesis that the intensificationof cropping systems by increasing C input and biodiversityunder NT restores SOC pool, increases resilience of degraded
agro-ecosystems, and enhances crop yield. The specific objec-tives of this research were to (i) assess the impact of thecontinuous plow-based CT on SOC stock for subtropicaland tropical agro-ecosystems vis-à-vis native vegetation(NV) as baseline; (ii) compare SOC balance among plow-based CT, NT cropping systems, and NV; and (iii) evaluatethe redistribution of SOC stock in the soil profile in relationto soil resilience and impact on the agronomic productivity.
MATERIALS AND METHODS
Sites
Field experiments were conducted at two agro-ecoregions inBrazil: (i) in a subtropical zone (25�090S–50�090W) locatednear the town of Ponta Grossa, in Paraná State, SouthernBrazil, at the experimental station of Instituto Agronômicodo Paraná (IAPAR), and hereafter called the PG site; and(ii) in a humid tropical zone (13�000S–55�580S) near thetown of Lucas do Rio Verde, in Mato Grosso State, centerwestern Brazil, at the experimental station of FundaçãoRio Verde, and hereafter called the LRV site (Figure 1). Adetailed description of location, climate, soil type, andchemical analyses of the two research sites is presentedin Table I.
Site Description, Land Uses Management, andExperimental Design
Ponta Grossa siteNative vegetation prior to the conversion to agricultural landconsisted of subtropical “prairies” dominated by C4 species(i.e., some fire resistant grasses such as Andropogon sp.,Aristida sp., Paspalum sp., and Panicum sp.) and by sub-tropical gallery forests, generally located in natural drainagechannels or on the summit landscape position (Maack,1981). The landscape has long gentle slopes ranging from2% to 7%. The soil of the experimental site is developedfrom the clastic sediments of the Devonian period character-ized by a mixture of Ponta Grossa shale. The soil has a deepand very well-structured profile, high porosity, and verygood internal drainage.
Table I. Sites description: altitude, soil type, parent material, climate, land use and management, duration of experiment, sampling depth,chemical properties, and clay content
Description Ponta Grossa, PR (PG site) Lucas do Rio Verde, MT (LRV site)
Altitude (m) 865 380Soil type (FAO) Red Latosol, Oxisol Red Yellow Latosol, OxisolSoil type (soil taxonomy) Rhodic Hapludox Typic HaplustoxParent material Shale Shale and sandstoneClimate type Mesothermic, summer and winter wet,
cold winter (Cfb)Humid tropic, summer hot and very wet,winter hot and dry (Aw)
Mean annual temperature (�C) 18�5 25�2Mean annual rainfall (mm) 1,545 1,950Land use NV, CT, MT, and NT NV, CT, and NT1–NT6Years of experiment 29 8Sampling depth (cm) 0–5, 5–10, 10–20, 20–40, 40–60,
60–80, and 80–1000–5, 5–10, 10–20, 20–40, 40–60,60–80, and 80–100
0–20 cm 593 681 637 619 393 381 39820–40 cm 673 713 690 667 437 425 44440–100 cm 717 731 719 692 529 455 490
NV, native vegetation of forest (PG) and Cerrado (LRV); FAO, Food and Agriculture Organization; CT, conventional plow-based tillage; MT, minimumtillage; NT, no-till. NT systems at the LRV site are described in Table III.†Mean values of the six NT systems were used for the quantification of the chemical properties and clay content.
CARBON DEPLETION BY PLOWING AND ITS RESTORATION BY NO-TILL SYSTEMS
In 1967, some of the NV area was converted to pasture-land by plowing to a 20-cm depth and then disking twiceto break the clods. In 1978, after 11 years of pasture, thisarea was converted to cropland and cultivated for 3 years:rice (Oryza sativa L.) for two and soybean (Glycine max,L. Merril) for one. The present experiment was initiated in1981, and the experimental design consisted of three tillagetreatments involving CT, minimum tillage (MT) and NT laidout as whole plots (Figure 2a). The CT consisted of plowingwith a 70-cm disk twice a year (i.e., after summer harvestand after winter harvest) to a 20-cm depth followed by two60-cm harrowing to break the clods for a uniform seedbed.The MT consisted on the use of one chisel plowing to a25-cm depth followed by one 60-cm narrow disking. In con-trast to that in CT and MT, there was no soil disturbance inNT. At each plot under NT, the dimension was 100� 100m,and for the CT and MT treatments, the dimension was50� 140m with six subplots (33� 50m for NT and25� 46m for MT and CT). The crop sequence used for overthe 10-year period for all tillage treatments comprised by a3-year cropping sequence, with two crops per year (detailsin Table II) with soybean (G.max, L. Merril) grown in sixand maize in four summers alternating with oats (Avenastrigosa Schreb), wheat (Triticum aestivum L.), and vetch(Vicia sativa L.) in the winter. An adjacent area to the experi-mental plots under NV was selected as a baseline to assess themanagement-induced changes in SOC stock. Six subplots weredemarcated for soil sampling on an area of 100� 140m under
NV. The mineral composition of the soil is similar among plotsand comprises 21�7% SiO2, 36�4% Al2O3, 13�1% Fe2O3, and2�1% TiO2. The average concentration of Fe, Si, and Al oxidesdid not vary among treatments or depths (Goncalves et al.,2008), and the soil of all plots has a similar textural composi-tion (Table I).
Lucas do Rio Verde siteThe NV, prior to its conversion to agricultural land, wasthe Cerrado forest comprising species about 8–20m tall(Sclerophillous and Xeromorphic). The climate of the regionis characterized by a humid season (Table I) between Octoberand April, followed by a pronounced dry season betweenMayand September.In 1986, the NVwas converted to agricultural land by clear-
ing, burning, and plowing. Immediately after the conversionof the NV between 1988 and 2000, soil acidity was managedusing dolomite lime incorporating 4�5Mgha�1 to a 20-cmdepth. After the latter, rice was cultivated for 1 year, andsoybean by ten consecutive years under plow-based tillage,involving a disk harrow to a 20-cm depth followed by a fallowperiod (Figure 2b). In 2001, an area of 10�9 ha was demarcatedto set up the present experiment.The experiment was conducted to assess the impact of
different cropping systems under NT on the basis of diversebiomass-C input, with the standard tillage system used inthe region comprised by one crop per year as monocultureunder CT in the summer (i.e., soybean alternating with
LAND DEGRADATION & DEVELOPMENT, (2013)
Figure 2. Chronology of land use in the experimental area: (a) Ponta Grossasite and (b) Lucas do Rio Verde site.
J. C. DE MORAES SÁ ET AL.
cotton – Gossypium hirsutum L.) were assigned as wholeplots. The tillage systems comprised seven treatments: (i)CT represented by the standard tillage systems used in theregion and comprised by an annual monoculture alternatingsoybean with cotton, and hereafter designated CT-S/Ct; (ii)
Table II. Site, tillage systems, crop sequence, and cumulative and annuaand 8 years experiment period at the Lucas do Rio Verde site
CT, plow-based conventional tillage; MT, minimum tillage; NT, no-till; S, soybea(Zea mays L.); W, wheat (Triticum aestivum L.); V, vetch (Vicia sativa); Ct, cot(Oriza sativa); Cs, Crotalaria spectabilis Roth; Fm, finger millet (Eleusine coraca(Helianthus annuus L.); Sg, Sorghum bicolor (L.) Moench; Mt, millet (Penniseexperiment at the Ponta Grossa (PG) site and 8 years at the Lucas do Rio Verde (Land LRV sites, respectively.†Crop sequence for the last 10 years at the PG site and crop sequence and tillage{C input by the crop residues.
NT1, comprised by the sequence of soybean as the first cropin the summer followed by maize (Zea mays L.) +Brachiariaruziziensis as the second crop; (iii) NT2, soybean as the firstcrop in the summer followed by finger millet (Eleusinecoracana) or finger millet + pigeon pea (Cajanus cajan) asthe second crop; (iv) NT3, soybean as the first crop in thesummer followed by a mix of finger millet + pigeon pea orfinger millet +Crotalaria spectabilis as the second crop; (v)NT4, soybean as the first crop in the summer followed by amix of finger millet +C. spectabilis or sunflower (Helianthusannuus) +B. ruziziensis as the second crop; (vi) NT5, soybeanas the first crop in the summer followed by a mix of sorghum(Sorghum bicolor) +B. ruziziensis as the second crop; and(vii) NT6, soybean as the first crop in the summer followedby millet (Pennisetum glaucum) or maize +B. ruziziensis asthe second crop. The details of the cropping sequences andbiomass-C input during the experimental period are presentedin Table II (for more details, see Tivet et al., 2013).Each plot under the NT treatments consisted of 216� 42m
with three subplots (72� 42m), and each plot under theCT treatment consisted of 216� 252m with four subplots(216� 62m). An adjacent area to the experimental plotsunder NV (i.e., Cerrado) of 200� 200m was selected as abaseline, and six subplots were demarcated for soil sampling.In average, the mineral compositions for the experimental areaconsisted of 23�3% SiO2, 39�0% Al2O3, 9% Fe2O3, and 2�2%TiO2. The clay content of all treatments was uniform (Table I).Further, there is no relationship between SOC concentrationand clay content in soil among all treatments (i.e., CerradoNV, CT, and diverse biomass-C input under NT).The soybean yield was computed for CT and all NT
treatments as an average among the subplots during 8 yearsof experiment. For CT, the soybean yield and the standard
l C input in the 29 years experiment period at the Ponta Grossa site
/M+Brz – M – S/Cs 60�8 7�60– M – S/M+Cs 58�0 7�25S/Fm+Cs – M – S/G+Cs 54�7 6�84/G +Brz – M – S/Sg +Cs 58�7 7�34S/Sg +Brz – M –S/Sg +Brz 67�0 8�38– M – S/M 59�3 7�41n (Glycine max (L.) Merr.); O, black oats (Avena strigosa Schreb); M, maizeton (Gossypium hirsutum L.); Brz, Brachiaria ruziziensis cv. ruzi; Rc, ricena (L.) Gaertn); Pp, pigeon pea (Cajanus cajan (L.) Mill sp.); G, sunflowertum typhoides Burm.); cumulative, sum of C input along the 29 years ofRV) site; annual C input, cumulative C input/29 and 8 years, for the PG site
systems for the 8 years experiment at the LRV site.
LAND DEGRADATION & DEVELOPMENT, (2013)
CARBON DEPLETION BY PLOWING AND ITS RESTORATION BY NO-TILL SYSTEMS
deviation in parenthesis, was 2,800� 280 kg ha�1, for NT1was 3,220� 204 kg ha�1, for NT2 was 3,270� 244 kg ha�1,for NT3 was 3,216� 173 kg ha�1, for NT4 was 3,218� 190kg ha�1, for NT5 was 3,323� 181 kg ha�1, and for NT6 was3,368� 226 kg ha�1.
Total Biomass (Aboveground and Belowground) andC Input
The annual inputs of C, from crop residues (roots and shoots)for each tillage treatment and summer crop, were calculatedon the basis of the harvest index and root and shoot ratio(Sá et al., 2001) during the experimental period (Table II).The aboveground biomass of the second crop (oat, vetch, B.ruziziensis, finger millet, pigeon pea, and C. spectabilis) wasmeasured during first year of the experiment, and the below-ground biomass was estimated from the previous studies(Sá et al., 2001). The concentration of C in the crop residues(i.e., gC kg�1 of dry matter), measured by the dry combustionmethod using an elemental CN analyzer (TruSpec CN,LECO, St. Joseph, MI, USA), was 395 for soybean, 450 forwheat, 455 for maize, 432 for oat, 444 for sorghum, 428 formillet, 385 for finger millet, 440 for crotalaria, 410 for pigeonpea, 412 for sunflower, and 443 g kg�1 for B. ruziziensis.
Soil Sampling
Soil samples at the end of the 29 years (September 2009) forthe PG site and 8 years (October 2009) of experimentationfor the LRV site were collected from seven depths: 0–5,5–10, 10–20, 20–40, 40–60, 60–80, and 80–100 cm. Bulksoil samples for each subplot were obtained for the depthsof 0–5, 5–10, and 10–20 cm by digging 15� 15-cm trenches.Soil samples for the depths of 20–40, 40–60, 60–80, and80–100 cm were obtained with an auger (4�5-cm diameter)from the same trench. Soil samples were obtained from fivepoints for each subplot, and composited. Soil bulk density
Table III. Soil bulk density (rb) of each site and for each sampling layer uor no-till (NT), and under the neighboring native vegetation (NV) at the PNV at the Lucas do Rio Verde site
Tillagesystem 0–5 5–10 10–20
PG site (Mgm�3)NV 1�04ns 0�95b 0�97bCT 1�01 1�05a 1�10aMT 1�02 1�10a 1�08aNT 1�02 1�10a 1�10aLRV site (Mgm�3)NV 1�00c 1�05c 1�13cCT 1�20b 1�31a 1�39abNT1 1�29ab 1�19b 1�27bNT2 1�30ab 1�23ab 1�30bNT3 1�25ab 1�30ab 1�32abNT4 1�21b 1�19b 1�25bNT5 1�32ab 1�29ab 1�31abNT6 1�37a 1�34a 1�28bLowercase letters indicate difference among land uses and tillage treatments. ns, msignificant difference was observed among tillage treatments [plow-based convenGrossa (PG) site and among the NT systems (NT1–NT6) systems at the Lucas do
(rb) was measured by the core method (Blake & Hartge,1986) using steel cylinders of 5� 5 cm (Table III). Cores from10- to 20-cm to 80- to 100-cm layers were taken in the middleof each layer. Bulk samples were oven-dried at 40 �C, gentlyground, sieved through a 2-mm sieve, and mixed.
Soil Chemical Properties andParticle Size-distribution Analyses
Soil pH was measured in 1:1 soil : water suspension(EMBRAPA, 1997), and exchangeable cations (Al3+, Ca2+,Mg2+, and K+) and available P were extracted for the 0- to20 cm and 20- to 40-cm layers using a cation and anionexchange resin (Raij & Quaggio, 1983). The cation exchangecapacity was obtained by the summation of exchangeable cat-ions and H+ (EMBRAPA, 1997). Soil textural analyses wereperformed for each subplot for the depths of 0–20, 20–40,and 40–100 cm (Table I) using the hydrometer with aBouyoucos scale (Gee & Bauder, 1986).
Total Organic Carbon Concentration, Stocks, andResilience Index
Subsamples of<2-mmbulk soil were finely ground (<150mm)for measuring SOC and total N concentrations by the dry com-bustion method using an elemental CN analyzer (TruSpecCN, LECO, St Joseph, MI, USA). The SOC stocks wereestimated to the 1-m depth and computed on an equivalent soilmass–depth basis (Ellert & Bettany, 1995).The rates of change of SOC (Mg ha�1 y�1) among NV
and CT, and among NT and CT, were estimated by usingEquations 1 and 2:
depletion rate ¼ SOCNV � SOCCTð Þ=t (1)
recovery rate ¼ SOCNT � SOCCTð Þ=t (2)
Where: SOCNV, SOCNT, and SOCCT refer to C stock underNV, NT, and CT, respectively, and t is the time (years) since
nder plow-based conventional tillage (CT), minimum tillage (MT),onta Grossa site, and under CT, NT cropping systems, and Cerrado
1�14c 1�10d 1�07cd 1�12b1�35a 1�27a 1�23a 1�25a1�29ab 1�11cd 1�13bcd 1�09b1�30ab 1�18bc 1�15b 1�10b1�32ab 1�16bcd 1�14bc 1�11b1�29ab 1�16bcd 1�07d 1�13b1�25b 1�15bcd 1�13bcd 1�09b1�27b 1�20b 1�13bcd 1�12beans within the same column are not significantly different at p< 0�05. Notional tillage (CT), minimum tillage (MT), and no-till (NT)] at the PontaRio Verde (LRV) site.
LAND DEGRADATION & DEVELOPMENT, (2013)
J. C. DE MORAES SÁ ET AL.
the conversion from NV to CT and from CT to NT. The resil-ience index (RI) was computed (Herrick & Wander, 1997;Dieckow et al., 2009) to assess the rate of SOC recovery fordifferent NT systems. This index uses NV as the upper limitand CT as the lower limit of SOC levels (Equation 3):
RI ¼ SOCNT � SOCCTð Þ= SOCNV � SOCCTð Þ (3)
The amount of C converted from crop residues to SOC(CCCRSOC) was calculated as follows:
CCCRSOC ¼ recovery rate=annual C inputð Þ � 100 (4)
Thus, CCCRSOC represents the percentage of C in thecrop residues converted into the SOC pool.
Statistical Analyses
Differences among treatments for SOC concentration, stock,and bulk density were tested through analysis of variance(ANOVA). Mean values were compared using the leastsignificant differences at the 5% probability level (Webster,2007). An analysis was performed by soil depth, and resultswere considered statistically significant at p< 0�05. Allstatistical calculations were carried out using R version 2.11.1(2006), package aov. SigmaPlot 12�0 (Systat Software Inc.,San Jose, California, USA) was used for graphic representation.Regression analyses were carried out to assess the relationshipin two ways: (i) the impact of annual C input on C sequesteredand (ii) the impact of SOC stock and soil resilience inagronomic productivity as measured by grain yield.
RESULTS
Depletion of Soil Organic Carbon and Total Nitrogen uponContinuous Plow-based Conventional Tillage and ItsRestoration by C Input through Cropping Systems under No-till
Ponta Grossa siteConcentrations of SOC and N decreased significantly with theincrease in soil depth and exhibited different patterns of distri-butions in the surface layer among land uses. These concentra-tions were highly stratified with depth under NV (Table IV)and were 80�7 gCkg�1 and 5�2 gNkg�1 in the 0- to 5-cmlayer, decreasing sharply to 30�6 and 2�06 gCkg�1 in the10- to 20-cm depth and 16�8 gCkg�1 and 0�88 gNkg�1 inthe 80–100 depth, respectively. However, there were nosignificant differences in SOC and N concentrations amongtillage treatments and the NV for the 10- to 100-cm depth.Further, a decreasing trend in SOC concentration was ob-served under MT and NV in comparison with that under NTor CT. Among tillage treatments, greater concentrations ofSOC and N, respectively, were recorded in soil under NTwith27�5 and 20�9 g kg�1 more SOC and 2�86 and 1�96 g kg�1
more N than those under CT and MT in the 0- to 5-cm depth.These data represent a decrease of 44% and 33% in SOC and60% and 41% inN, respectively, in CT andMT than in the NTsystems. Expectedly, the SOC concentration under NV, CT,MT, and NT was positively correlated with N concentration
(SOC=11�4N+10.5, R2 = 0�86, n=188, p< 0�0001) acrossall soil depths and indicate a C :N ratio of 11:4.Measurements of rb and concentrations of SOC under
NV in an adjacent undisturbed soil to a 1-m depth (Tables IIIand IV) indicated that the SOC stock in the 0- to 20-cmdepth represented 36% of the total compared with only27% under CT SOC stock (Table V). Further, land-usechanges significantly impacted on SOC and N stocks(Table V). For example, the average SOC stock in the 0- to20-cm depth decreased from 92�0Mgha�1 in NV (Table V)to 67�4Mgha�1 in CT plot (p< 0�001), a decline of ~27%over 42 years since the conversion of NV into cultivated fieldand the use of CT for >29 years. This decline represents aloss of ~0�58MgCha�1 y�1 and ~62�0 kgNha�1 y�1 forthe clayey Rhodic Hapludox. The soil under NT in the 0- to20-cm depth contained 17�0 and 14�3Mgha�1 more SOCand 2�02 and 1�2Mgha�1 more N than those under CT andMT (Figure 3 and Table V). These data represent an annualaccumulation rate of 0�59 and 0�49MgCha�1 for SOC and69�7 and 41�2 kg ha�1 of N over 29 years under CT and MT,respectively. The rate of SOC sequestration is relatively lowcompared with that in previous results reported by Sá et al.(2001) in an Oxisol near this experiment and with that inAmado et al. (2006) in a similar subtropical region. The mainreason was the low amount of biomass-C added through resi-dues of summer crop and winter species during the 29 years ofexperiment and the near absence of legumes as a winter covercrop (Table II). Recently, Ferreira et al. (2012) reported thatfor this region, the minimum amount of biomass-C inputto maintain the steady state was 3�21Mg ha�1 y�1, and inthis study, the biomass-C input was 4�15Mg ha�1 y�1. Thepercentage of biomass-C converted into SOC in this studywas 14�2 and was very similar with the findings obtainedby Ferreira et al. (2012), which was ~12�63–14�26%.There were no significant differences among tillage treat-ments even with a trend of soils under CT and NTcontaining relatively more SOC than that under MT in the40- to 100-cm depth.
Lucas do Rio Verde siteConcentrations of SOC and N decreased with the increase indepth and exhibited differential distribution in the soil profileamong the land-use treatments (Table IV). Concentrations ofSOC and N, respectively, under NV were 38�3 gCkg�1
and 1�6 gN kg�1 in the 0- to 5-cm depth, compared with7�1 gC kg�1 and 0�34 gN kg�1 in the 80- to 100-cm depth.Similarly, SOC and N concentrations under the NT systemsalso declined with depth, but the stratification was less pro-nounced than that under the Cerrado NV. The SOC concen-tration under cropped fields (CT and NT) was significantlylower (p< 0�05) than that under NV in the 0- to 5-cmdepth. Total N concentration was likewise higher underthe Cerrado NV than that under CT and NT, but no differ-ence among land uses was observed below the 40-cmdepth. The soil under NT, with a predominance of cereal assecond crops [maize +B. ruziziensis, sorghum+B. ruziziensis,and millet] during the dry season, and higher biomass input
LAND DEGRADATION & DEVELOPMENT, (2013)
TableIV
.Soilo
rganiccarbon
andtotaln
itrogen
concentrations
(gkg
�1)managed
underplow
-based
conventio
naltillage(CT),minim
umtillage,and
no-till
(NT),atthePontaGrossasite,and
undertheCTandNTcropping
system
s(N
T1–
NT6)
attheLucas
doRio
Verde
site,and
undertheneighboringnativ
evegetatio
nforboth
sites
Landuse
Soildepth(cm)
0–5
5–10
10–20
20–40
40–60
60–80
80–100
SOC(gkg
�1)
PG
site
NV
80�7
4�9{
a†43
�54�9
ns30
�62�0
ns28
�92�2
ns22�6
2�0
ns18
�51�0
b16
�81�7
nsCT
35�0
0�6
c35
�01�0
33�8
0�6
28�5
1�1
24�2
1�5
21�8
1�2
a20
�21�2
MT
41�7
1�7
c40
�61�1
30�3
1�0
26�4
1�0
21�8
0�7
19�8
0�7
ab18
�50�4
NT
62�5
1�4
b44
�31�0
32�1
0�4
28�5
0�5
24�1
0�4
21�8
0�6
a20
�30�7
p-values
<0�0
001
0�079
0�161
0�554
0�519
0�046
0�135
LRV
site
NV
38�3
3�9
a21
�31�1
ns16
�11�0
ns12
�20�6
a9�5
0�5
ns8�3
0�6
ns7�1
0�2
nsCT
18�3
0�4
b17
�50�5
13�4
0�6
9�50�1
c7�8
0�1
7�10�2
7�10�4
NT1
25�2
2�5
b20
�81�4
18�2
1�5
11�0
0�5
abc
8�20�3
7�50�4
7�00�2
NT2
21�5
1�6
b18
�81�1
16�7
0�6
11�4
0�6
abc
8�60�6
7�10�0
6�80�2
NT3
22�4
0�8
b17
�81�0
15�3
1�0
10�4
0�2
bc8�3
0�4
7�00�2
7�00�1
NT4
19�7
0�9
b18
�30�8
16�2
0�8
12�1
1�4
ab8�9
0�6
7�60�5
6�80�2
NT5
23�6
2�3
b20
�82�1
18�4
2�8
12�3
0�7
ab8�7
0�3
7�10�2
6�60�3
NT6
24�0
1�9
b18
�80�7
16�7
0�3
11�7
0�2
ab9�1
0�3
7�50�3
6�90�4
p-values
<0�0
001
0�132
0�174
0�053
0�188
0�416
0�909
TN
(gkg
�1)
PG
site
NV
5�20
0�29
a2�8
10�26
a2�0
60�08
ns1�7
00�17
ns1�1
30�22
ns1�0
00�18
ns0�8
80�19
nsCT
1�93
0�07
b1�8
40�14
b1�6
80�07
1�30
0�15
1�07
0�16
0�87
0�12
0�82
0�09
MT
2�83
0�43
b2�6
40�43
a1�6
80�31
1�44
0�25
1�23
0�21
1�25
0�20
1�16
0�18
NT
4�79
0�19
a2�9
90�12
a1�6
60�07
1�51
0�10
1�11
0�09
0�98
0�12
0�95
0�12
p-values
<0�0
001
0�029
0�302
0�436
0�912
0�318
0�317
LRV
site
NV
1�61
0�17
a1�0
60�09
ns0�7
60�06
ns0�5
90�06
ns0�4
90�09
ns0�3
70�05
ns0�3
40�03
bcCT
1�03
0�05
b0�8
60�07
0�62
0�07
0�39
0�03
0�37
0�03
0�38
0�02
0�35
0�02
abNT1
1�40
0�21
ab1�0
40�19
0�76
0�14
0�47
0�03
0�38
0�02
0�31
0�01
0�31
0�02
bcNT2
1�04
0�21
b0�9
50�14
0�73
0�06
0�41
0�03
0�37
0�02
0�31
0�02
0�27
0�02
cNT3
1�28
0�05
ab0�7
70�05
0�65
0�05
0�43
0�05
0�41
0�02
0�36
0�02
0�42
0�01
aNT4
0�93
0�05
b0�7
60�03
0�68
0�03
0�54
0�05
0�45
0�04
0�36
0�03
0�31
0�05
bcNT5
1�19
0�17
ab0�8
80�02
0�78
0�03
0�52
0�01
0�46
0�04
0�36
0�02
0�30
0�01
bcNT6
1�33
0�17
ab0�9
20�12
0�83
0�11
0�53
0�06
0�48
0�04
0�37
0�05
0�33
0�02
bcp-values
0�044
0�336
0�473
0�070
0�335
0�543
0�038
PG,P
onta
Grossa;
LRV,L
ucas
doRio
Verde;SOC,soilorganiccarbon;TN,total
nitrogen;NV,n
ativevegetatio
n;CT,p
low-based
conventio
naltillage;MT,m
inim
umtillage;NT,n
o-till;ns,n
otsignificant.
†Low
ercase
letters
referto
thecomparisonam
ongtillage
treatm
entswith
indepth.
{ Superscript
numbers
referto
thestandard
error.
CARBON DEPLETION BY PLOWING AND ITS RESTORATION BY NO-TILL SYSTEMS
Table V. Effects of land use and management practices on soil organic C and total nitrogen stocks (Mg ha�1) at the Ponta Grossa and Lucasdo Rio Verde sites for the depths of 0–20, 20–40, 40–60, 60–80, and 80–100 cm, and under the neighboring native vegetation for both sites
p-values 0�236 0�612 0�995 0�448 0�431PG, Ponta Grossa; LRV, Lucas do Rio Verde; SOC, soil organic carbon; TN, total nitrogen; NV, native vegetation; CT, plow-based conventional tillage; MT,minimum tillage; NT, no-till; ns, not significant.†Lowercase letters refer to the comparison among tillage treatments within depth.{Superscript numbers refer to the standard error.
J. C. DE MORAES SÁ ET AL.
(NT1, NT5, and NT6), contained on average higher SOC andN concentrations in the 0- to 5-cm depth than those in the soilunder CT (23�6–25�2 vs. 18�3 gCkg�1, and 1�2–1�4 vs. 0�92 gNkg�1). The SOC concentration in soil under NV andcropped fields is positively correlated with the N concentra-tion (SOC=19�3N+ 1�25, R2 = 0�90, n=188, p< 0�001)across all soil depths and indicates a C :N ratio of 19:3.Soilrbwas lowest in the 0- to 5-cm layer (1�00� 0�1Mgm�3)
under NV, but it increased slightly to 1�12� 0�02Mgm�3 inthe 1-m depth (Table III). It was higher in some crop lands(NT and CT) in the 0- to 20-cm depth, probably because oftillage (CT) and traction (seeding and harvesting) (Ferreiraet al., 2011). Soil rb under CT in the 40- to 100-cm depthwas significantly more than that under the Cerrado NV andsome NT treatments (p< 0�05).The SOC stock to the 1-m depth was 130�3Mg ha�1 in
the Cerrado NV, of which 36% was stored in 0- to 20-cm,23% in 20- to 40-cm, and 41% in 40- to 100-cm depth
(Table V). Significant losses of the SOC stock occurredupon its conversion to an agricultural land use and by along-term use of CT (Figure 4 and Table V). Consequently,the difference in the SOC stock between CT and the CerradoNV was 14�2 and 6�0MgC ha�1 in the 0- to 20-cm and20- to 40-cm depths, respectively. The difference in theSOC stock between CT and the Cerrado NV (20�1Mg Cha�1) in the 0- to 40-cm layer represented a 27% decreaseby 23 years of CT, with an average rate of loss of 0�88MgC ha�1 y�1 for the clayey Typic Haplustox. In contrast, theSOC stock in the 0- to 20-cm and 20- to 40-cm layersunder NT1, NT5, and NT6 was similar to that under theCerrado NV (Figure 4 and Table V).The SOC stocks in the 0- to 20-cm depth under NT, with
a predominance of grasses during the dry season (NT1,NT5, and NT6), were significantly higher (p< 0�05) comparedwith those under CT. The difference in the SOC stock in the20-cm depth between all NT systems andCT ranged on average
LAND DEGRADATION & DEVELOPMENT, (2013)
Figure 3. Effects of continuous plow-based conventional tillage (CT) and conversion from CT to no-till (NT) cropping systems with contrasting C inputs onsoil organic C (SOC) (Mg ha�1) CT and from CT to NT systems at the Ponta Grossa site. Clay and silt %, and annual C input are given per tillage treatment.The SOC resilience index ([SOCNT�SOCCT] / [SOCNV�SOCCT]) is given for the NT system for the 0- to 20-cm depth interval, where SOCNT is the C stockin the field under NT, SOCCT is the C stock in the field under CT, SOCNV is the C stock in the field under native vegetation. Error bars indicate standard errorsof the means. Uppercase letters indicate difference among land use and tillage treatments. The level of significance for each soil depth interval is indicated.
Figure 4. Effects of continuous plow-based conventional tillage (CT) and conversion from CT to no-till (NT) cropping systems with contrasting C inputson soil organic C (SOC) (Mgha�1) at the Lucas do Rio Verde site. Clay and silt %, and annual C input are given per land use. The SOC resilience index([SOCNT�SOCCT] / [SOCNV�SOCCT]) is given for each NT system for the 0- to 40-cm depth interval, where SOCNT is the C stock in the field underNT, SOCCT is the C stock in the field under CT, SOCNV is the C stock in the field under native vegetation. Error bars indicate standard errors of themeans. Uppercase letters indicate difference among land use and tillage treatments. The level of significance for each soil depth interval is indicated. Data
of the annual C input to soil under Cerrado NV are from Corbeels et al. (2006).
CARBON DEPLETION BY PLOWING AND ITS RESTORATION BY NO-TILL SYSTEMS
from 3�9 to 10�4MgCha�1, representing average sequestrationrates of 0�48–1�30MgCha�1 y�1. These rates of SOC seques-tration in the 0- to 40-cm depth increased from 0�73 (NT3) to1�98MgCha�1 y�1 (NT5) and represented an increase of52% more in the 20- to 40-cm layer. No significant differenceswere reported between any treatments in the 40- to 100-cmdepth, but the SOC stocks under CT and NT were lowerthan those under the Cerrado NV from 3�1 to 6�0MgC ha�1.
Enhancement of Soil Resilience
The RI was 0�69 under NT at the PG site and ranged from 0�29(NT3) to 0�79 (NT5) at the LRV site in accord with the in-crease in dry matter input among the NT systems (RI= 0�18
C input – 0�76, R2 = 0�88, p< 0�001) and indicating the poten-tial of biomass-C input under the NT systems to restore SOCdepleted by the conversion of the NV cropland managed byCT (Figures 3 and 4).
DISCUSSION
Soil Organic Carbon Restoration and Sequestration Ratesin Response to Cropping Systems under No-till
At the PG site, concentrations of SOC andNwere highly strat-ified with depth under NT, whereas those under CT and MTwere uniform in the 0- to 20-cm layer owing to the mixingeffect of successive plowing (Sá & Lal, 2009). The reduction
LAND DEGRADATION & DEVELOPMENT, (2013)
Figure 5. Relationship between soil organic C (SOC) gain (Mgha�1 y�1)and annual C input (Mgha�1 y�1) under the no-till (NT) systems for thedepths of 0–100, 0–20, 20–40, and 40–60 cm at the Lucas do Rio Verdesite. The annual C input, C sequestered (0–100 cm), and the % of C from
crop residues converted to SOC are given in the inserted table.
J. C. DE MORAES SÁ ET AL.
of SOC in the surface layers of soil under CT occurs through ahigh rate of oxidation process because of soil disturbance andthe illuviation of SOC into the subsoil (Osher et al., 2003).Although no significant differences were observed, soil underCT and NT stored 16�9 and 16�6Mgha�1 more SOC, respec-tively, than that under NV for the depths of 40–60, 60–80, and80–100 cm (Table V). The differences in the SOC stockamong tillage systems and NV at depth may be attributed tothe following: (i) more translocation of labile compoundsfrom surface to subsoil in arable land vis-à-vis NV byincreased volume of water moving through the soil profilein each rainfall event (White, 1985; Sollins & Radulovich,1988); (ii) high clay content in CT soil (Table I) contribut-ing to a higher adsorptive capacity of organic compoundson clay minerals (Feller & Beare, 1997); and (iii) possibledifferences in SOC stock under NV and cropland at the timeof initiating the experiment in 1989. In addition, the SOCstock in the subsoil may be affected by land-use changesand agricultural practices. Osher et al. (2003) reported thatsome of the C depleted in the topsoil under a sugarcane(Saccharum officinarum) plantation was translocated into thesubsoil and concluded that the loss of SOC upon land-useconversion can be overestimated if gains in the subsoil arenot considered. Determination of 13C natural abundance forNV and a tillage chronosequence, as well as the quantificationof stable and labile SOC fractions reported by Sá et al. (2001),confirm these assumptions.The rates of SOC sequestration in the present study under
NT cropping systems are in accord with those of severalothers studies in Brazil (Sá et al., 2001; Sisti et al., 2004;Bayer et al., 2006a; Boddey et al., 2010). In comparisonwith studies on soils of the temperate regions (Baker et al.,2007; Blanco-Canqui & Lal, 2008; Poirier et al., 2009),higher SOC stocks under CT than NT were not observedbelow the plow depth. Computed to the 1-m depth, thedifferences in the SOC stock among NT and CT are no morethan those computed to the 20-cm depth (Table V).At the LRV site, the SOC stocks at different depths in the
Cerrado NV are in accord with those of Batlle-Bayer et al.(2010). The SOC stocks to the 20-cm layer have beenreported at 54Mg ha�1 for a Typic Haplustox (Bayeret al., 2006a), 68�1Mg ha�1 for a Geri-Gibbsic Ferralsol(Corbeels et al., 2006), and 46–63Mg ha�1 for some RedLatosol (Chapuis-Lardy et al., 2002).The rate of the loss of the SOC stock under CT (0�88Mg
C ha�1 y�1 in the 0- to 40-cm depth) is in contrast with thatobserved by Batlle-Bayer et al. (2010) who reported the lossof the SOC stock in CT between �0�3 and +0�1MgCha�1
y�1. Zinn et al. (2005a) observed no significant changes inthe SOC stocks for some Oxisols containing either less clay(<500 g kg�1) or more clay (>500 g kg�1) for contrastingland-use systems in the Cerrado region. However, Silvaet al. (1994) reported losses of 41% (clayey soils), 76%(loamy soils), and 80% (sandy soils) of the original SOCstock after 5 years of heavy harrowing for the cultivation ofsoybean. Bayer et al. (2006a) reported a loss of 15% of SOCin a sandy clay loam (350 g clay kg�1) sloping land (12%)
managed under CT for more than 20 years. Neto et al.(2010) observed that land-use conversion induced a total lossof around 15MgCha�1 after 10 years of continuous CT.Similarly, Séguy et al. (2006) reported an average loss of1�6Mg C ha�1 y�1 following the conversion of the CerradoNV to soybean monoculture for 10 years in a soil located invicinity of the LRV site.As much as 32–37% of the total SOC stock to the 1-m
depth was present in the 0- to 20-cm layer under NTcropping systems. No significant differences in the SOCstock were observed among the NT systems, whereas adifference of almost 10�0MgCha�1 was observed betweenthe maximum and minimum SOC stocks. In addition, moreSOC stocks in the 0- to 40-cm depth (Table V) were observedin the soil under NT that received a high amount of drybiomass annually (NT1, NT5, and NT6). Trends of the differ-ences in the N stock among the NT systems were similar tothose observed for the SOC stock, with NT1, NT5, and NT6containing higher stocks than other system (Figure 4). The rateof SOC sequestration under some NT (i.e., NT5 and NT6) incomparison with CT increased significantly with the increasein soil depths. With reference to the SOC sequestration indifferent depths (Table V), approximately 65% of the seques-tration occurred in the 0- to 20-cm depth and the remainder(i.e., 35%) in the 20- to 40-cm depth. Vegetation types, incombination with climate and mineralogy, affect the verti-cal distribution of SOC (Jobbagy & Jackson, 2000), and
LAND DEGRADATION & DEVELOPMENT, (2013)
Figure 6. Relationship between (a) soil organic C (SOC) stock (x-axis, Mg ha�1) under the no-till (NT) systems for 0–100 cm and soybean yield (y-axis, kgha�1) and (b) the resilience index (RI) (x-axis) and yield (y-axis, kg ha�1) at the Lucas do Rio Verde site.
CARBON DEPLETION BY PLOWING AND ITS RESTORATION BY NO-TILL SYSTEMS
these NT systems have distinct imprints on the nature anddepth distribution of SOC. Although additional research iswarranted, SOC sequestration in the subsoil could be attrib-uted to the illuviation of dissolved organic C (Lorenz &Lal, 2005; Sá & Lal, 2009), higher SOC rhizodepositionin the soil profile by deep rooting systems of grains/foragegrasses (B. ruziziensis, sorghum, and finger millet) andlegumes (Crotalaria sp., C. cajan) (Fisher et al., 1994; Séguyet al., 2006), and enhancement of vertical distribution throughbioturbation (Wilkinson et al., 2009), which influences SOCdynamics (Lavelle et al., 2006). In addition, these speciesenhance root/arbuscular mycorrhizal association that enhanceexudates, increase soil aggregation (Wright & Upadhyaya,1998; Rillig, 2004), and accentuate the C allocation to thewhole root system (Jones et al., 1991), with a significantproportion diverted to the fungal component (Johnson et al.,2002). Additional research, including the determination of13C natural abundance for NV, CT, and NT soils, is neededto draw valid conclusions and to demonstrate that diversebiomass-C inputs under NT are efficient in terms of SOCsequestration even below the 20-cm depth. Such thematicresearch is essential to identifying cropping systems thatenhance the SOC stock, improve soil fertility, enrich thediversity and abundance of microflora and fauna communities(Brevault et al., 2007; Hungria et al., 2009), improve porosityand hydraulic conductivity (Cassaro et al., 2011), increaseaggregation (Madari et al., 2005), and strengthen nutrientcycling (Chapuis-Lardy et al., 2002). In addition, the SOCconcentration decreases sharply with depth (Table IV),whereas the clay content follows an opposite trend (Table I).The specific surface area and thus the adsorptive capacity ofminerals in the subsoil are higher than those in the surfacelayer. The interaction between clay + silt content and SOC(Zinn et al., 2005b) indicates an unfilled C sink capacity inthe subsoil, and that deep placement of SOC can contributeto higher rates of SOC sequestration.
Soil Resilience and Agronomic Productivity as a Function ofC Sequestration
In accord with some previous studies (Paustian et al., 1992;Bayer et al., 2000; Kong et al., 2005), this study also indicatesa strong linear relationship between annual C input (x-axis)
and annual SOC sequestration (SOC gain, y-axis) to the 1-mdepth (Figure 5). Although the slope of the function depictingrelative annual C input and annual C sequestration to the 1-mdepth is high, the fraction of C input by crop residuesconverted into SOC was ~20�5% and varied among croppingsystems (Table II and Figure 5). There also exits a strongrelationship between the SOC stock (1-m depth) and the RI(RI=0�044SOCstock + 4�62, R2 = 0�97, p< 0�001), indicatingthat the increase in SOC enhances soil quality (Figure 6aand b). Expectedly, therefore, the highest RI is associated withthe highest rate of SOC sequestration (NT5) and confirms thestrong relationship between these parameters. Thus, thebiomass-C needed to maintain a positive C balance at theLRV and PG sites is estimated at ~5�5 and ~4�0MgCha�1
y�1 (i.e., ~12�5 and~ 8�0Mgha�1 of dry matter), respectively.Relatively rapid changes in the SOC stock with intensive
NT systems indicate that there is a large potential to reversethe process of soil degradation and of SOC decline throughconversion to the NT systems on the basis of complex rota-tions and cover cropping. Moreover, the enhancement of soilresilience through SOC sequestration had a positive and lineareffect on the agronomic productivity (Figure 6a). Every 1-Mgincrease in the SOC stock to the 1-m depth increased soybeanyield by 28 kg ha�1, and every 0�1-unit increase in RIincreased soybean yield by 600 kg ha�1 (Figure 6b).
CONCLUSIONS
The data presented indicate that the SOC stock is drasticallyreduced by the conversion of NV to agro-ecosystem by acontinuous use of plow-based CT leading to a depletion of0�58 and 0�67MgCha�1 y�1 in the 0- to 20-cm depth atthe PG and LRV sites, respectively. Significant differencesin the SOC stocks to the 20-cm depth between CT and NTcropping systems, and the rate of SOC sequestration of0�59Mg C ha�1 y�1 in subtropical region and 0�48–1�30MgC ha�1 y�1 in the tropical Cerrado region, were observed.Any increase in SOC sequestration in the subsoil of CTtreatment was not observed at either of the two sites. In NT,however, SOC accumulation increased with the increase inthe input of biomass-C at the LRV site. A high SOC resilienceunder the tropical NT systems indicates a considerable potential
LAND DEGRADATION & DEVELOPMENT, (2013)
J. C. DE MORAES SÁ ET AL.
to reverse the process of soil degradation and SOC decline byconversion to intensive NT systems (high and diversifiedannual C input). The results support the hypothesis that SOCrestoration is determined by the intensification of the NTcropping systems through the continuous input of biomass-Cto maintain the C flow in the soil supplying the process of soilC transformations and driving the production systems.
ACKNOWLEDGEMENTS
We thank the Agricultural Research Institute of Paraná andthe Lucas do Rio Verde Foundation for allowing access tothe experimental fields. We greatly appreciate the helpfrom Mrs. Jaqueline Aparecida Gonçalves and Mr. RomeuMartins Filho for laboratory analyses. Authors thank theAgrisus Foundation (project number: PA 677/10), Centrede Coopération Internationale en Recherche Agronomiquepour le Développement and the Food and Agriculture Organi-zation for the financial support.
REFERENCES
Amado TJ, Bayer C, Conceicao PC, Spagnollo E, Costa de Campos B-H, daVeiga M. 2006. Potential of carbon accumulation in no-till soils withintensive use and cover crops in southern Brazil. Journal of EnvironmentalQuality 35: 1599–1607.
Baker JM, Ochsner TE, Venterea RT, Griffis TJ. 2007. Tillage and soilcarbon sequestration – what do we really know? Agriculture, Ecosystemsand Environment 118: 1–5.
Batlle-Bayer L, Batjes NH, Bindraban PS. 2010. Changes in organic carbonstocks upon land use conversion in the Brazilian Cerrado: a review.Agriculture, Ecosystems and Environment 137: 47–58.
Bayer C, Lovato T, Dieckow J, Zanatta JA, Mielniczuk J. 2006a. A methodfor estimating coefficients of soil organic matter dynamics based on long-term experiments. Soil and Tillage Research 91: 217–226.
Bayer C, Martin-Neto L, Mielniczuk J, Ceretta CA. 2000. Effect of no-tillcropping systems on soil organic matter in a sandy clay loam acrisol fromSouthern Brazil monitored by electron spin resonance and nuclearmagnetic resonance. Soil and Tillage Research 53: 95–104.
Bayer C, Martin-Neto L, Mielniczuk J, Pavinato A, Dieckow J. 2006b.Carbon sequestration in two Brazilian Cerrado soils under no-till. Soiland Tillage Research 86: 237–245.
Bayer C, Martin-Neto L, Mielniczuk J, Pillon CN, Sangoi L. 2001. Changesin soil organic matter fractions under subtropical no-till cropping systems.Soil Science Society of America Journal 65: 1473–1478.
Bernoux M, Cerri CC, Cerri CEP, Neto MS, Metay A, Perrin AS, Scopel E,Razafimbelo T, Blavet D, Piccolo MD, Pavei M, Milne E. 2006.Cropping systems, carbon sequestration and erosion in Brazil, a review.Agronomy for Sustainable Development 26: 1–8.
Blake GR, Hartge KH. 1986. Bulk density. InMethods of soil analysis, partI. Physical and mineralogical methods: agronomy monograph no. 9 (2nded.), Klute A (ed). ASA SSSA: Madison, WI, USA; 363–375.
Blanco-Canqui H, Lal R. 2008. No-tillage and soil-profile carbon seques-tration: an on-farm assessment. Soil Science Society of America Journal72: 693–701.
Boddey RM, Jantalia CP, Conceicao PC, Zanatta JA, Bayer C, MielniczukJ, Dieckow J, Dos Santos HP, Denardin JE, Aita C, Giacomini SJ,Alves BJR, Urquiaga S. 2010. Carbon accumulation at depth inFerralsols under zero-till subtropical agriculture. Global Change Biology16: 784–795.
Brevault T, Bikay S, Maldes JM, Naudin K. 2007. Impact of a no-till withmulch soil management strategy on soil macrofauna communities in acotton cropping system. Soil and Tillage Research 97: 140–149.
Broecker WS. 1975. Climatic change – are we on brink of a pronouncedglobal warming. Science 189: 460–463.
Carvalho JLN, Cerri CEP, Feigl BJ, Piccolo MC, Godinho VP, Cerri CC.2009. Carbon sequestration in agricultural soils in the Cerrado regionof the Brazilian Amazon. Soil and Tillage Research 103: 342–349.
Cassaro FAM, Borkowski AK, Pires LF, Rosa JA, Saab SD. 2011. Character-ization of a Brazilian clayey soil submitted to conventional and no-tillagemanagement practices using pore size distribution analysis. Soil and TillageResearch 111: 175–179.
Cerri CEP, Sparovek G, Bernoux M, Easterling WE, Melillo JM, Cerri CC.2007. Tropical agriculture and global warming: impacts and mitigationoptions. Scientia Agricola 64: 83–99.
Chapuis-Lardy L, Brossard M, Assad MLL, Laurent JY. 2002. Carbonand phosphorus stocks of clayey Ferralsols in Cerrado native andagroecosystems, Brazil. Agriculture, Ecosystems and Environment92: 147–158.
Corbeels M, Scopel E, Cardoso A, Bernoux M, Douzet JM, Neto MS.2006. Soil carbon storage potential of direct seeding mulch-basedcropping systems in the Cerrados of Brazil. Global Change Biology12: 1773–1787.
Denef K, Six J, Merckx R, Paustian K. 2004. Carbon sequestration inmicroaggregates of no-tillage soils with different clay mineralogy. SoilScience Society of America Journal 68: 1935–1944.
Dick WA. 1983. Organic-carbon, nitrogen, and phosphorus concentrationsand pH in soil profiles as affected by tillage intensity. Soil Science Societyof America Journal 47: 102–107.
Dieckow J, Bayer C, Conceicao PC, Zanatta JA, Martin-Neto L, Milori DBM,Salton JC,MacedoMM,Mielniczuk J, Hernani LC. 2009. Land use, tillage,texture and organic matter stock and composition in tropical and subtropicalBrazilian soils. European Journal of Soil Science 60: 240–249.
Dieckow J, Mielniczuk J, Knicker H, Bayer C, Dick DP, Kogel-Knabner I.2005. Composition of organic matter in a subtropical acrisol as influencedby land use, cropping and N fertilization, assessed by CPMAS C-13NMR spectroscopy. European Journal of Soil Science 56: 705–715.
Ellert BH, Bettany JR. 1995. Calculation of organic matter and nutrientsstored in soils under contrasting management regimes. Canadian Journalof Soil Science 75: 529-538.
EMBRAPA. 1997. Manual de métodos de análise de solo. 2ed. CentroNacional de Pesquisa de Solos (Rio de Janeiro, RJ): Rio de Janeiro; 390 pp.
Feller C, Beare MH. 1997. Physical control of soil organic matter dynamicsin the tropics. Geoderma 79: 69–116.
Ferreira AO, Sá JCM, Balarezo Giarola NF, Harms MG, Miara S, BavosoMA, Briedis C, Netto CQ. 2011. Variation in aggregate-tensile strengthas a function of carbon content of two soils in the region of CamposGerais. Revista Brasileira de Ciência do Solo 35: 437–445.
Ferreira AO, Sá JCM, Harms MG, Miara S, Briedis C, Netto CQ, BurknerJS, Canalli LB. 2012. Carbon balance and crop residue management indynamic equilibrium under a no-till system in Campos Gerais. RevistaBrasileira de Ciência do Solo 36: 1583–1590.
Fisher MJ, Rao IM, Ayarza MA, Lascano CE, Sanz JI, Thomas RJ, VeraRR. 1994. Carbon storage by introduced deep-rooted grasses in theSouth-American savannas. Nature 371: 236–238.
Gee GW, Bauder JW. 1986. Particle-size analysis. In Methods of soilanalysis, Klute A (ed). ASA and SSSA: Madison, WI, USA; 383–411.
Goncalves D, Leite WC, Brinatti AM, Saab SD, Iarosz KC, MasearenhasYP, Carneiro PIB, Rosa JA. 2008. Mineralogy of a Red Latosol underdifferent management systems for twenty-four years. Revista Brasileirade Ciência do Solo 32: 2647–2652.
Herrick JE, Wander MM. 1997. Relationships between soil organiccarbon and soil quality in cropped and rangeland soils: the importanceof distribution, composition, and soil biological activity. CRC Press: BocaRaton, FL.
Hungria M, Franchini JC, Brandao-Junior O, Kaschuk G, Souza RA. 2009.Soil microbial activity and crop sustainability in a long-term experimentwith three soil-tillage and two crop-rotation systems. Applied Soil Ecol-ogy 42: 288–296.
Jantalia CP, Resck DVS, Alves BJR, Zotarelli L, Urquiaga S, Boddey RM.2007. Tillage effect on C stocks of a clayey Oxisol under a soybean-basedcrop rotation in the Brazilian Cerrado region. Soil and Tillage Research95: 97–109.
Jobbagy EG, Jackson RB. 2000. The vertical distribution of soil organiccarbon and its relation to climate and vegetation. Ecological Applications10: 423–436.
Johnson D, Leake JR, Ostle N, Ineson P, Read DJ. 2002. In situ (CO2)-C-13pulse-labelling of upland grassland demonstrates a rapid pathway ofcarbon flux from arbuscular mycorrhizal mycelia to the soil. NewPhytologist 153: 327–334.
Jones MD, Durall DM, Tinker PB. 1991. Fluxes of carbon and phosphorusbetween symbionts in willow ectomycorrhizas and their changes withtime. New Phytologist 119: 99–106.
LAND DEGRADATION & DEVELOPMENT, (2013)
CARBON DEPLETION BY PLOWING AND ITS RESTORATION BY NO-TILL SYSTEMS
Kern JS, Johnson MG. 1993. Conservation tillage impacts on national soiland atmospheric carbon levels. Soil Science Society of America Journal57: 200–210.
Kong AYY, Six J, Bryant DC, Denison RF, van Kessel C. 2005. Therelationship between carbon input, aggregation, and soil organic carbonstabilization in sustainable cropping systems. Soil Science Society ofAmerica Journal 69: 1078–1085.
Lal R. 2004. Soil carbon sequestration impacts on global climate changeand food security. Science 304: 1623–1627.
Lal R. 2008. Carbon sequestration. Philosophical Transactions of the RoyalSociety B-Biological Sciences 363: 815–830.
Lavelle P, Decaens T, Aubert M, Barot S, Blouin M, Bureau F, Margerie P,Mora P, Rossi JP. 2006. Soil invertebrates and ecosystem services.European Journal of Soil Biology 42: S3–S15.
Lorenz K, Lal R. 2005. The depth distribution of soil organic carbon inrelation to land use and management and the potential of carbon seques-tration in subsoil horizons. In Advances in agronomy, Vol 88. ElsevierAcademic Press Inc: San Diego; 35–66.
Maack R. 1981. Classificação do clima do Estado do Paraná, Livraria JoséOlímpio Editora S.A., Rio de Janeiro, Brazil. 450 pp.
Madari B, Machado P, Torres E, de Andrade AG, Valencia LIO. 2005. Notillage and crop rotation effects on soil aggregation and organic carbon ina Rhodic Ferralsol from Southern Brazil. Soil and Tillage Research80: 185–200.
Neto MS, Scopel E, Corbeels M, Cardoso AN, Douzet J-M, Feller C,Piccolo MdC, Cerri CC, Bernoux M. 2010. Soil carbon stocks underno-tillage mulch-based cropping systems in the Brazilian Cerrado:an on-farm synchronic assessment. Soil and Tillage Research 110:187–195.
Ogle SM, Swan A, Paustian K. 2012. No-till management impacts on cropproductivity, carbon input and soil carbon sequestration. Agriculture,Ecosystems and Environment 149: 37–49.
Osher LJ, Matson PA, Amundson R. 2003. Effect of land use change on soilcarbon in Hawaii. Biogeochemistry 65: 213–232.
Paustian K, Andren O, Janzen HH, Lal R, Smith P, Tian G, Tiessen H, VanNoordwijk M, Woomer PL. 1997. Agricultural soils as a sink to mitigateCO2 emissions. Soil Use and Management 13: 230–244.
Paustian K, Parton WJ, Persson J. 1992. Modeling soil organic-matter inorganic-amended and nitrogen-fertilized long-term plots. Soil ScienceSociety of America Journal 56: 476–488.
Poirier V, Angers DA, Rochette P, Chantigny MH, Ziadi N, Tremblay G,Fortin J. 2009. Interactive effects of tillage and mineral fertilizationon soil carbon profiles. Soil Science Society of America Journal73: 255–261.
R Development Core Team. 2006. R: a language and environment forstatistical computing. R Foundation for Statistical Computing: Vienna,Austria. ISBN 3- 900051-07-0.
Raij Bv, Quaggio JA. 1983. Métodos de análises de solo para fins defertilidade. Campinas, Instituto Agronômico, 31pp.
Rasmussen PE, Allmaras RR, Rohde CR, Roager NC. 1980. Crop residueinfluences on soil carbon and nitrogen in a wheat-fallow system. SoilScience Society of America Journal 44: 596–600.
Rillig MC. 2004. Arbuscular mycorrhizae, glomalin, and soil aggregation.Canadian Journal of Soil Science 84: 355–363.
Sá JCD, Cerri CC, Dick WA, Lal R, Venske SP, Piccolo MC, Feigl BE.2001. Organic matter dynamics and carbon sequestration rates for atillage chronosequence in a Brazilian Oxisol. Soil Science Society ofAmerica Journal 65: 1486–1499.
Sá JCD, Lal R. 2009. Stratification ratio of soil organic matter pools as anindicator of carbon sequestration in a tillage chronosequence on a BrazilianOxisol. Soil and Tillage Research 103: 46–56.
Séguy L, Bouzinac S, Husson O. 2006. Direct-seeded tropical soil systemswith permanent soil cover: learning from Brazilian experience. In
Biological approach to sustainable soil systems, Uphoff N, Ball AS,Fernandes E et al. (eds). CRC Press, Taylor and Francis: New York,USA; 323–342.
Silva JE, Lemainski J, Resck DVS. 1994. Perdas de matéria orgânica esuas relações com a capacidade de troca catiônica em solos da regiãode cerrados do oeste baiano. Revista Brasileira de Ciência do Solo18: 541–547.
Sisti CPJ, dos Santos HP, Kohhann R, Alves BJR, Urquiaga S, BoddeyRM. 2004. Change in carbon and nitrogen stocks in soil under 13 yearsof conventional or zero tillage in southern Brazil. Soil and TillageResearch 76: 39–58.
Six J, Conant RT, Paul EA, Paustian K. 2002a. Stabilization mechanisms ofsoil organic matter: implications for C-saturation of soils. Plant and Soil241: 155–176.
Six J, Feller C, Denef K, Ogle SM, Sa JCD, Albrecht A. 2002b. Soil organicmatter, biota and aggregation in temperate and tropical soils – effects ofno-tillage. Agronomie 22: 755–775.
Six J, Frey SD, Thiet RK, Batten KM. 2006. Bacterial and fungal contribu-tions to carbon sequestration in agroecosystems. Soil Science Society ofAmerica Journal 70: 555–569.
Sollins P, Radulovich R. 1988. Effects of soil physical structure on solutetransport in a weathered tropical soil. Soil Science Society of AmericaJournal 52: 1168–1173.
Tivet FE, Sá JCM, Lal R, Briedis C, Borszowskei PR, Santos JB, Farias A,Eurich G, Hartman DC, Nadolny M, Jr. 2013. Aggregate C depletion byplowing and its restoration by diverse biomass-C inputs under no-till insub-tropical and tropical regions of Brazil. Soil and Tillage Research126: 203–218.
Uphoff N, Ball AS, Fernandes E, Herren H, Husson O, Palm C, Pretty J,Sanginga N, Thies J. 2006. Understanding the functioning and manage-ment of soil systems. In Biological approaches to sustainable soilsystems, Uphoff N, Ball AS, Fernandes E, Herren H, Husson O, LaingM, Palm C, Pretty J, Sanchez P, Sanginga N, Thies J (eds). CRC Press,Taylor and Francis: New York, USA; 3–13.
Varvel GE, Wilhelm WW. 2011. No-tillage increases soil profile carbonand nitrogen under long-term rainfed cropping systems. Soil and TillageResearch 114: 28–36.
Virto I, Barre P, Burlot A, Chenu C. 2012. Carbon input differences as themain factor explaining the variability in soil organic C storage in no-tilledcompared to inversion tilled agrosystems. Biogeochemistry 108: 17–26.
Webster R. 2007. Analysis of variance, inference, multiple comparisons andsampling effects in soil research (vol 58, pg 74, 2007). European Journalof Soil Science 58: 1549–1549.
West TO, Post WM. 2002. Soil organic carbon sequestration rates by tillageand crop rotation: a global data analysis. Soil Science Society of AmericaJournal 66: 1930–1946.
West TO, Six J. 2007. Considering the influence of sequestration durationand carbon saturation on estimates of soil carbon capacity. ClimaticChange 80: 25–41.
White RE. 1985. The influence of macropores on transport of dissolved andsuspended matter through soil. In: Science AiS, Stewart BA (ed).Springer: New York, N.Y; 327 pp.
Wilkinson MT, Richards PJ, Humphreys GS. 2009. Breaking ground:pedological, geological, and ecological implications of soil bioturbation.Earth-Science Reviews 97: 257–272.
Wright SF, Upadhyaya A. 1998. A survey of soils for aggregate stabilityand glomalin, a glycoprotein produced by hyphae of arbuscular mycorrhizalfungi. Plant and Soil 198: 97–107.
Zinn YL, Lal R, Resck DVS. 2005a. Changes in soil organic carbon stocksunder agriculture in Brazil. Soil and Tillage Research 84: 28–40.
Zinn YL, Lal R, Resck DVS. 2005b. Texture and organic carbon relationsdescribed by a profile pedotransfer function for Brazilian Cerrado soils.Geoderma 127: 168–173.
