Printer Friendly

Soil fertility, physical and chemical organic matter fractions, natural [sup.13]C and [sup.15]N abundance in biogenic and physicogenic aggregates in areas under different land use systems.

Introduction

The relationship among the activity of soil biota, soil organic matter (SOM) and soil aggregate dynamics has been identified and intensively studied since the early 1900s (Six et al. 2004; Bronick and Lai 2005; Rillig and Mummey 2006). Biological, physical and chemical cycles fuse unitary particles, forming microaggregates (Tisdall and Oades 1982; Amezketa 1999). In a study on the dynamics of aggregate formation and stabilisation, Tivet et al. (2013) showed that aggregates lose carbon (C) when native forests are converted into conventional tillage systems (CTS), managed with ploughing and harrowing. Those authors also found that soil C can be later recovered by converting CTS to a no-till system (NTS). They further showed that CTS stops the formation of new soil aggregates by dispersing particles of clay or silt + clay microaggregates, but with NTS, new aggregates can be formed and C redistributed among them by incorporating different plant residues.

Aggregation promotes the proper functioning of soil processes because it affects water flow, infiltration, aeration, nutrient retention, microbial activity, C sequestration and greenhouse gas emission, and because aggregates are the habitat of soil organisms (Six el al. 2000; Barthes and Roose 2002; Bronick and Lai 2005; Costa et al. 2012). Soil aggregates, particularly those >2 mm, are also good indicators of soil quality (Vezzani and Mielniczuk 2011; Loss et al. 2011; Briedis et al. 2012; Tivet et al. 2013).

Different systems of land use can determine specific changes in soil attributes, thereby affecting aggregation and ultimately modifying the formation of biogenic aggregates. Studies that attempt to explain qualitative and quantitative aspects of aggregate formation as a function of soil management are scarce, possibly because of the complexity and difficulty of establishing references that allow identification of the processes of aggregate formation (Velasquez et al. 2007; Jouquet et al. 2009; Cecillon et al. 2010; Batista et al. 2013). According to these studies, soil aggregates can be classified into biogenic and physicogenic, based on their morphology. Round-shaped biogenic aggregates are produced by soil fauna, especially earthworms (Oligochaeta) and/or root-associated microorganisms. Peres et al. (1998) identified, in soil images, biological macropores produced by earthworm activity and associated their size and shape with earthworm species and stage of development. Physicogenic aggregates are angular or prism-shaped as a result of wetting and drying cycles.

Biogenic aggregates contain a higher amount of macro- and micronutrients than surrounding soil (Langenbach et al. 2002; Fiuza et al. 2011; Jouquet et al. 2011; Batista et al. 2013). These differences can result from earthworm activity, because these animals are estimated to ingest 200-400 Mg soil [ha.sup.-1] [year.sup.-1] (Lavelle et al. 1983). According to the source material and quality of soil organic matter (SOM), the biogenic aggregates produced can improve soil fertility. However, few studies have addressed aggregates of biogenic origin, possibly because of difficulties related to their identification, cycling time and position in the soil matrix (Brussaard et al. 2007; Velasquez et al. 2007).

Soil aggregation can provide physical protection against rapid decomposition of SOM caused by soil management (Pulleman and Marinissen 2004), which along with soil type affects SOM amount and quality. Therefore, changes to soil environment caused by agricultural management can affect SOM production, thereby modifying the route for biogenic and physicogenic aggregate formation (Pulleman 2002).

The separation between biogenic and physicogenic aggregates has been successfully used as an indicator of soil quality in a study that compared earthworm-made soil aggregates and the surrounding soil environment (Velasquez et al. 2007; Jouquet et al. 2011). Different land management systems (e.g. no-till, conventional till, pasture and forest) affect not only the formation of SOM fractions but also the C content of these fractions in biogenic and physicogenic aggregates (Pulleman and Marinissen 2004; Cecillon et al. 2010; Batista et al. 2013). As a result, both SOM and aggregate formation contribute to soil fertility and therefore to soil quality.

Based on work of Velasquez et al. (2007) and Jouquet et al. (2011), the present study tested the hypothesis that biogenic aggregates are more sensitive to modifications imposed by different soil management systems and are therefore better indicators of soil quality than physicogenic aggregates. We classified soil aggregates from Marmeleiro, Parana State, Brazil into biogenic and physicogenic based on formation processes and morphological features; characterised the soil exchange complex (pH in water-saturated soil paste, [Ca.sup.2+]; [Mg.sup.2+], [Al.sup.3+], [K.sup.+], [Na.sup.+], [H.sup.+] + [Al.sup.3] + and P) of the biogenic and physicogenic aggregates; quantified the distribution of total organic C (TOC) and nitrogen (N) levels and natural [sup.13]C and [sup.15]N abundance in the biogenic and physicogenic aggregates; and determined C levels in oxidisable fractions, particle-size fractions and humic fractions of SOM in biogenic and physicogenic aggregates.

Materials and methods

Location, climate and soil in the study area

The study was performed on a rural property in Marmeleiro, south-west Parana State, Brazil (Fig. 1). Climate in the area is subtropical (Koppen Cfa climate), with well-defined seasons, characterised by mild winters and hot summers. Rainfall is well distributed throughout the year. Soil is classified as Alfisol (Soil Survey Staff 2010) with clayey texture. According to the Sistema Brasileiro dc Classificacao do Solo (Embrapa 2006), the soil is Nitossolo Vermelho.

The following areas under different soil management systems were evaluated (Fig. 1): NTS for 15 years, CTS for 56 years, and secondary forest (Forest) and pasture grass (carpetgrass, Axonopus compressus) (Pasture) each established for >30 years. The original area was composed of mixed ombrophilous forest. The NTS area (26[degrees]14'43.3"S, 53[degrees]10'20.4"W; 753 m altitude) was cropped with successive soybean (Glycine max)-ryegrass (Lolium multiflorum) planting. The CTS area (26[degrees]14'53.6"S, 53[degrees]10'24.6"W; 740 m altitude) had been managed with ploughing and harrowing for 56 years and always planted with maize (Zea mays); in the last 14 years it was planted with tobacco (Nicotiana tabacum) as main crop and maize as off-season crop. After the maize harvest, it was sown with black oats (Avena strigosa) as winter herbage for dairy cattle. The Forest (26[degrees]14'35.0"S, 53[degrees]10'17.3"W; 747 m altitude) and Pasture (26[degrees]14'59.1"S, 53[degrees]10'30.3"W: 713 m altitude) areas surrounding the croplands were considered as references, representing the original soil features. In the Pasture area, dairy cattle grazed in an extensive system, at an approximate stocking density of 1.4 animal unit [ha.sup.-1] (1 animal unit = 1 cow of 450 kg liveweight).

The NTS area was fertilised with 290 kg [ha.sup.-1] of N - [P.sub.2][O.sub.5] - [K.sub.2]O (0:18:18 formulation) at soybean sowing and limed every 5-6 years with 1240 kg [ha.sup.-1] of limestone. Ryegrass was sowed in March and remained until October, when soybean was sown over ryegrass straw. The CTS area was fertilised with 850 kg [ha.sup.-1] of N - [P.sub.2][O.sub.5] - [K.sub.2]O (10: 18:20 formulation) at tobacco planting and received 400 kg [ha.sup.-1] of urea as topdressing. When undisturbed soil samples were collected, the NTS area was covered by ryegrass, whereas the CTS area had been ploughed and harrowed 2 weeks before.

Sample collection and investigation of aggregate formation

The areas under the different land use systems exhibited similar topography, soil conditions and climate, with only NTS and CTS differing in relation to the practices adopted (crop rotation and soil tillage). As such, nearly 1.0 ha (100m by 100 m) was demarcated and undisturbed samples were collected in four trenches opened across the sowing lines. In the Pasture area, cattle grazing sites were avoided. In Forest, which was adjacent to the other areas studied, soil was sampled among the trees, in sites under the influence of the root system and in the central portion of the area. The samples were taken from the 0-5 and 5-10 cm layers, and a composite sample was formed with three undisturbed samples from each layer, with four replicates per area. The samples were identified, stored in plastic bags, and taken to the laboratory, where they were air-dried and sieved through 9.7- and 8.0-mm mesh to separate the aggregates.

The process of aggregate formation was investigated in the sieved aggregates (8.0-9.7 mm). To determine relative mass contribution, l00g of aggregate was weighted for each replicate and area. As proposed by Bullock et al. (1985), the aggregates were observed under magnifier and manually separated into physicogenic and biogenic according to their morphological patterns. Biogenic aggregates, produced by soil- and/or root-associated microorganisms, were round. Physicogenic aggregates, formed by wetting and drying cycles, showed angular or prism shapes (Velasquez et al. 2007; Jouquet et al. 2009).

Sample analyses

For chemical analysis, physicogenic and biogenic aggregates were crumbled and sieved through 2.0-mm mesh. This material was used in the following analyses.

Characterisation of aggregate exchange complex

The pH (of saturated soil paste) and calcium ([Ca.sup.2+]), magnesium ([Mg.sup.2+]), aluminium ([Al.sup.3+]), potassium ([K.sup.+]), sodium ([Na.sup.+]), hydrogen ([H.sup.+]) + [Al.sup.3+], and phosphorus (P) were determined according to the Brazilian Agricultural Research Corporation (Embrapa 1997). The pH in IUO was determined by potentiometer, using a 1 :2.5 (v/v) soil: solution. Exchangeable cations [Ca.sup.2+], [Mg.sup.2+] and [Al.sup.3+] were extracted with I m KCl, and potential soil acidity (H + Al) was determined using a solution of 0.5 m calcium acetate, at pH 7.0. Phosphorus, [Na.sup.+] and [K.sup.+] were extracted with 0.0125m [H.sub.2]S[O.sub.4] solution + 0.05m HCl. Contents of [Ca.sup.2+] and [Mg.sup.2+] were determined by titration with 0.0125 m EDTA solution, Na and K by flame photometry, P by colourimetry, and [Al.sup.3+] and H + Al by titration with 0.025 M NaOH.

Carbon and nitrogen levels

The TOC and total N (TN) levels in soil aggregates were determined by the dry combustion method, in a C and N auto-analyser at 900[degrees]C (LECO CHN-1000; LECO Corporation, St Joseph, MI, USA) at the Laboratory of Isotopic Ecology of the Center for Nuclear Energy in Agriculture, Piracicaba, Sao Paulo State, Brazil. Because no carbonate-derived C is found in Alfisols, total C was referred to as organic C (Schumacher 2002).

According to Loss et al. (2014), the C:N ratio in bulk soil (0-5 cm) (g C: g N [kg.sup.-1] soil) from NTS, CTS, Forest and Pasture areas was 43.4:3.4, 19.4:2.0, 38.0:3.1 and 39.1:4.0, respectively. In the 5-10cm layer, the C:N ratio was 29.7:2.3, 18.7:2.0, 27.5 :2.2 and 37.9:2.0, respectively.

Origin of carbon and nitrogen in the aggregates

The content of [sup.15]N and [sup.13]C was determined by mass spectrophotometry (DELTAplus; Finnigan MAT, Bremen, Germany) at the abovementioned Laboratory of Isotopic Ecology.

Chemical partitioning of aggregate SOM

The humic substances were separated into fulvic acid fraction (FAF), humic acid fraction (HAF) and humin (HUM) via the differential solubility technique established by the International Society of Humic Substances (Swift 1996) and adapted by Benites et al. (2003). Air-dried, ground soil (1.0 g) was incubated with 20 mL of 0.1 M NaOH solution for 24 h. Separation of the alkaline extract (i.e. C-FAF + C-HAF) and the residue (C-HUM) was achieved by centrifugation at 5000g for 30 min. The tubes were washed with the same NaOH solution, obtaining a final volume of ~40mL. The residue was then removed from the centrifuge tubes, placed in a Petri dish and dried at 65[degrees]C until reaching constant weight. The pH of the alkaline extract was adjusted to 1.0 ([+ or -]0.1) with 20% [H.sub.2]S[O.sub.4]. The solution was decanted in a refrigerator for 18 h, and the precipitate (C-HAF) was separated from the solution (C-FAF) by filtration; distilled water was added to the solution to a final volume of 50 mL. Organic carbon quantification in the fractions (C-HUM, C-FAF and C-HAF) was performed according to Yeomans and Bremner (1988).

Granulometric partitioning of aggregate SOM

Particulate organic C (POC) and mineral-associated organic C (MOC) were quantified according to Cambardclla and Elliott (1992). Approximately 20 g of air-dried, ground soil was mixed with 60 mL of a sodium hexametaphosphate solution (5 g L 1) and stirred for 15 h with a horizontal stirrer. The suspension was passed through a 53-[micro]m sieve forced by water jet. The material retained on the sieve corresponded to sand + POC and was oven-dried at 60[degrees]C for relative weight quantification. Dried samples were ground in a ceramic mortar and TOC was determined according to Yeomans and Bremner (1988). Particles that passed through the 53-pm sieve corresponded to MOC from silt and clay fractions. MOC was calculated by the difference between TOC and POC.

Oxidisable organic carbon fraction in the aggregates

Partitioning of oxidisable carbon was performed by decreasing the oxidation degree (lability) according to Chan et al. (2001). In an adaptation of the Walkley-Black method, four oxidisable fractions were analysed: very easily oxidisable (FI), easily oxidisable (F2), moderately oxidisable (F3), and resistant (F4). The four quantified fractions corresponded to carbon oxidation with [K.sub.2][Cr.sub.2][O.sub.7] acidified with the following concentrations of [H.sub.2]S[O.sub.4]: 3 m (F1), 6-3 m (F2), 9-6 m (F3), and 12-9 m (F4).

Statistical analyses

After confirming normality with the Lilliefors test and homoscedasticity with Cochran's and Bartlett's tests, data were analysed for a completely randomised design involving four systems (NTS, CTS, Forest, Pasture) with four replications. The different land-use systems were under similar topography, soil and climate conditions, differing only in relation to soil management and plant cover. The results were subjected to analysis of variance (C-test), and statistically different means were contrasted by least significant difference (l.s.d.(-Student's t-test, at a significance level of P=0.05. Statistical analyses were carried out to compare land-use systems (NTS, CTS, SF, PG) and aggregate types (biogenic and physicogenic). Statistical analyses were performed using the SAEG software, version 9.1 (SAEG 2007).

Results and discussion

Relative contribution of the different aggregates

In both layers evaluated (0-5 and 5-10 cm), physicogenic aggregates dominated over biogenic aggregates (Table 1). This finding corroborates the results of Batista et al. (2013), who reported that physicogenic aggregates were more abundant than biogenic and intermediate aggregates in a crop-livestock integration area in the state of Mato Grosso do Sul, Brazil. In the CTS area, only physicogenic aggregates were found, probably because it was ploughed and harrowed a few days before soil sampling. This indicates that CTS does not favour the formation and maintenance of biogenic aggregates, especially compared with the other systems evaluated (Table 1).

Similar to our findings in the CTS, Pulleman et al. (2005) found that 67% of soil mass in the 0-10 cm layer corresponded to physicogenic aggregates and only 7.4% to biogenic aggregates in a CTS area established for 70 years. The lack of biogenic aggregates in the CTS area decreases SOM levels because biogenic aggregates contain greater amounts of SOM than physicogenic aggregates and promote SOM stabilisation (Brussaard et al. 2007).

The proportion of biogenic and physicogenic aggregates was similar in NTS and Forest areas, regardless of soil layer (Table 1). These results indicate that the environment produced in NTS was favourable to the formation and/or maintenance of biogenic aggregates, in contrast to CTS. Plant straw on the soil surface and the absence of soil turnover improve the physical and chemical conditions of soil, with a subsequent increase in soil fauna, especially earthworms (Anghinoni et al. 2011). Thus, the formation of biogenic aggregates increases since it is directly related to soil macrofauna (Cecillon et al. 2010; Batista et al. 2013).

In the Pasture area, biogenic aggregates were found in greater amounts than in NTS and Forest (Table 1), probably because of the dynamic root system of the broadleaf carpetgrass. This forage grass promotes addition of C to soil, thereby increasing soil fauna and producing aggregates of biogenic origin. The higher content of biogenic aggregates in the Pasture area was indeed accompanied by higher levels of C (Table 2). Similar results were reported by Pulleman et al. (2005), who found greater amounts of biogenic than physicogenic aggregates in Pasture areas and high TOC levels in biogenic aggregates in the 0-10 cm layer.

Fertility, C and N content, and isotope composition of aggregates

Mean pH values (in saturated soil paste) did not differ between aggregate types in the study areas. The lowest pH values in both aggregate types were found in the Pasture area (in the two layers), which is directly related to the higher potential soil acidity (H + Al) in this area. In addition, biogenic and physicogenic aggregates from the Pasture area exhibited higher Al levels at 5-10 cm depth. In the 0-5 cm layer, H + AI was lower in biogenic than physicogenic aggregates from NTS and lower in physicogenic than biogenic aggregates from Forest. No differences were found between aggregates in Pasture in the 0-5 or 5-10 cm layer (Table 2).

Exchangeable Ca levels in both aggregate types were generally higher in Forest than in the other areas, although in the 5-10 cm layer, Ca content in physicogenic aggregates was similar between NTS and Forest. The biogenic aggregates of highest Ca levels were found in Forest areas at both 0-5 cm and 5-10 cm. For physicogenic aggregates, Ca levels were lower in Pasture and CTS than in NTS and Forest areas. Magnesium levels were lower in physicogenic aggregates from CTS than from other areas at 0-5 cm, and in both aggregate types from the Pasture area at 5-10 cm. Only in Forest did biogenic aggregates have significantly more Mg than physicogenic aggregates (Table 2).

Soil fertility in Rio de Janeiro state, Brazil, is considered high according to the study of Freirc et al. (2013), who obtained Ca + Mg >6.10 [cmol.sub.c] [dm.sup.-3] by the Embrapa method (Embrapa 1997). Therefore, the areas evaluated in the present study, especially the Forest area, showed high Ca + Mg levels in both biogenic and physicogenic aggregates. The only exception was the Pasture area at 5-10 cm. The high Ca + Mg content results from the high natural fertility of the predominant soil, an Alfisol. This soil is derived from basaltic rock and it may contain interstratified kaolinite-smectite, which according to Teske et al. (2013), increases the cation exchange capacity (CEC). As reported by those authors, interstratification accounts for the high CEC of basalt-derived soils containing smectite in southern Brazil.

The NTS and Pasture areas showed lower Ca levels than Forest, probably because Ca uptake by crops and pasture grass is significant. The Pasture area was not constantly grazed, and reshooting grass (after being eaten by cattle) would have required water and nutrients such as Ca and Mg. Because of the low anthropogenic interference, the Forest area showed higher nutrient cycling and lower nutrient loss. The low Ca and Mg content detected in the CTS area (compared with NTS and Forest areas) indicates that disturbance of soil aggregates by ploughing and harrowing increases SOM mineralisation, thereby lowering TOC levels (Table 2). As a result, the capacity for cation retention decreases (especially because of the low SOM content), lowering Ca and Mg levels (Table 2).

The higher Ca and Mg levels in physicogenic aggregates from NTS and Forest areas than from the CTS area (Table 2) indicate that NTS is more efficient than CTS in maintaining and even increasing Ca and Mg content, probably because crops are rotated and straw applied as protective cover (mulching) and soil is not turned over.

Comparing the physicogcnic aggregates from NTS and CTS areas, the former exhibited higher Ca and Mg levels at 0-5 and 5-10 cm and higher K., N and TOC levels at 0-5 cm (Table 2). The conservation practices adopted in NTS contribute to maintaining soil moisture and avoiding direct exposure of the soil surface to sunlight, decreasing surface temperature and, consequently, SOM mineralisation (Loss et al. 2009). NTS was therefore more efficient in increasing soil nutrient levels than CTS.

Biogenic aggregates in the Forest area (at both depths) exhibited the highest Ca content. A likely explanation is earthworm castings, which during aggregate separation were found to be more frequent in this area. In the Pasture area, root activity was the main factor affecting biogenic aggregate formation. Earlier studies indicated that areas containing more biogenic aggregates derived from earthworm castings are richer in nutrients such as Ca than areas containing physicogcnic aggregates (Silva Neto et al. 2010). The calcifcrous glands of earthworms likely contribute to worm casting enrichment during food digestion (Schrader and Zhang 1997).

In both layers studied, levels of K were similar between aggregate types within the same land-use system. However, K levels were higher in Pasture, at both depths. As reported by Cruscio and Borghi (2007), pasture grasses show large capacity for recycling nutrients through their root system, which can uptake nutrients at depth. Those authors also attribute this result to straw incorporation, which, along with the large and deep root system, contributes to the efficiency of grasses in using nutrients. At 0-5 cm depth, the lowest K levels were found in biogenic aggregates from NTS and physicogenic aggregates from CTS (Table 2).

Phosphorus levels were higher in biogenic aggregates in all land-use systems and soil layers studied. This result is directly related to the higher TOC levels of biogenic aggregates (Table 2), corroborating earlier studies by Silva Neto et al. (2010) and Batista et al. (2013). Those authors found higher P levels in biogenic aggregates derived from earthworm castings and associated with root activity. The lowest P levels were found in Forest (0-5 and 5-10 cm) and Pasture (5-10 cm) areas, probably because, in contrast to the cropped areas (NTS and CTS), these sites were not fertilised.

There were no significant differences in N content between biogenic and physicogenic aggregates within land-use systems, except in the Forest area (at 0-5 cm), where biogenic aggregates had higher N content than physicogenic aggregates. This finding is associated with litter deposition in Forest and higher TOC levels in biogenic aggregates from this site. At 0-5 cm depth, the lowest N content for biogenic aggregates was found in NTS, and for physicogenic aggregates in CTS, indicating N mineralisation in cropped areas. In the 5-10 cm layer, the highest N levels within both aggregate types were also found in the Forest area (Table 2). As observed with K., levels of N were lower in CTS, indicating that ploughing and harrowing not only disrupt soil aggregates but also expose the SOM (and TOC) that was protected inside the aggregates, impairing the routes of biogenic aggregate formation (Table 1). Consequently, very mobile nutrients such as N and K. are lost through leaching. In the CTS area, the low levels of these nutrients were associated with low TOC levels.

In fact, physicogenic aggregates from CTS did contain the lowest TOC levels, in both soil layers. By contrast, the highest TOC levels were found in the two aggregate types in the Pasture area (at both depths) and in physicogenic aggregates from Forest at 5-10 cm. The carpetgrass cover explains the high TOC levels in Pasture, given that it produces a large amount of plant matter with high C : N ratio (Table 3) and has a root system that is efficient in exploring at depth. This was observed in the biogenic aggregates of the Pasture area, which, when separated under magnifier, were found to be associated with several roots. Biogenic aggregates from Forest were less associated with roots and showed more worm castings than in the other areas.

At both depths evaluated, biogenic aggregates exhibited higher TOC levels than their physicogenic counterparts, which agrees with earlier studies (Bossuyt et al. 2005; Pulleman et al. 2005; Jouquet et al. 2009; Silva Neto et al. 2010; Batista et al. 2013). According to Bossuyt et al. (2005), TOC content in biogenic aggregates from worm castings was 22% higher than in physicogenic aggregates. In the present study, the high TOC content of biogenic aggregates was directly related to the high Ca, Mg, N, P and FI + Al levels in Forest and the high P levels in NTS, Forest and Pasture areas, similar to the report by Fiuza et al. (2011), who also found higher TOC and nutrient content in biogenic aggregates and argued that TOC of worm-casting-derived biogenic aggregates was positively correlated with P, K, Ca and Mg levels in forest, rubber plantation and pasture environments.

Lower isotope [sup.13]C content was found in Forest, NTS and CTS areas than in Pasture, indicating the predominance of [C.sub.3] plants. As with other forest regions in Brazil, the Forest area showed very low [sup.13]C levels, nearly -26%c at both depths (Table 3). Cropped areas showed [sup.13]C values between those of [C.sub.3] and [C.sub.4] plants. In CTS, the intermediate isotopic signal reflected the combination of tobacco and oat ([C.sub.3] plants) with com ([C.sub.4] plant), and in NTS, where both crops were C3 plants, it reflected spontaneous growth of [C.sub.4] plants in the area. A characteristic isotopic 13C value for [C.sub.4] plants was found in the Pasture area, ranging from ~-16[per thousand] (at 0-5 cm) to -17.89[per thousand] for biogenic and -18.12[per thousand] for physicogenic aggregates at 5-10cm (Table 3).

Isotopic enrichment at depth (from 0-5 to 5-10 cm) was observed primarily in the Pasture area, possibly because SOM decomposition and humification processes release higher levels of [sup.12]C at the soil surface, and 'old' organic matter becomes more enriched in [sup.13]C than recently incorporated matter (O'Brien and Stout 1978; Vitorello et al. 1989; Martin et al. 1990). The CTS area, however, exhibited the same isotope signal across the layers, illustrating that ploughing and harrowing homogenise the tillable portion of the soil, which is deeper than the 10 cm evaluated.

Biogenic aggregates from Forest and NTS areas contained lower [sup.13]C levels, whereas those from Pasture contained higher levels of this isotope, irrespective of soil layer. These results are related to the high TOC content of biogenic aggregates and to the isotopic signal of the predominant vegetation in the evaluated areas. Accordingly, Jouquet et al. (2009) found lower [sup.13]C values and higher TOC content in biogenic aggregates. Those authors suggest that [sup.13]C and [sup.15]N content indicates the ecological position or the feeding strategy of earthworms, thereby serving as a tool to differentiate biogenic from physicogenic aggregates.

The highest [sup.15]N values in biogenic aggregates were found in the Pasture area (in both layers). In physicogenic aggregates, the highest [sup.15]N values were also observed in the Pasture area at 0-5 cm, and in the Pasture and CTS areas at 5-10 cm. In the Pasture area, the absence of legumes and the constant deposition of cattle manure likely increased soil [sup.15]N. Manures are typically more enriched in [sup.15]N than soil (Yoneyama 1996), and the deposition of this material increases soil l 5N (Szpak 2014).

The high [sup.15]N levels in the Pasture area can also be attributed to the high chemical stability of its SOM. The most humified SOM fractions usually show higher [sup.15]N levels than poorly decomposed forms (Natelhoffer and Fry 1988). As such, the Pasture area showed a higher MOC content (Fig. 2), which is the most recalcitrant SOM fraction. MOC is associated with the clay fraction of soil (Cambardclla and Elliott 1992), and in uncultivated soils, the fine clay contains residual N derived from microbial biomass turnover and higher content of humified compounds that are enriched in [sup.15]N (Szpak 2014).

The Forest area showed the lowest [sup.15]N levels in the two layers and aggregate types (Table 3). According to Yoneyama (1996) and Szpak (2014), forest soils that are not disturbed by human activity become more impoverished in [sup.15]N than agricultural soils. In forests without N-based fertilisation and with low biological [N.sub.2] fixation, rain is the only N source in the system. In addition, a balance between N inputs and outputs due to internal N cycling slightly increases soil [sup.15]N (Szpak 2014).

Biogenic and physicogenic aggregates from the NTS area showed higher [sup.15]N values than those from the Forest area, and at 0-5 cm depth, physicogenic aggregates from the NTS area contained lower [sup.15]N levels than those from the CTS area. In the NTS area, N is obtained from biological [N.sub.2] fixation (by soybean crop) and chemical fertilisation. In the CTS area, chemical fertilisers were the only N source, and they had been extensively applied for a long period (56 years), whereas in the NTS area they had been applied for 15 years. Soil N input by biological fixation produces lower [sup.15]N levels than obtained with CTS (at 0-5 cm). Therefore, soils with continuous cultivation of [N.sub.2]-fixing legumes (e.g. soybean in the NTS area) likely exhibit lower [sup.15]N levels than those cropped with non-fixing species (e.g. the CTS area, without legumes) (Szpak 2014).

Only in the Forest and NTS areas were [sup.15]N levels different between the aggregates, with the higher levels found in physicogenic aggregates (Table 3). The difference can be explained by the predominance of organic N forms in biogenic aggregates and mineral N in physicogenic aggregates. Mineral N forms (ammonium, nitrate) show rapid changes, from hours to days, in the [sup.15]N ; [sup.14]N ratio, whereas for organic N forms (amino acid, proteins, humus), these changes can take months or years to occur (Karamanos and Rennie 1978, 1981). Therefore, the higher the content of N bound to organic shapes, the lower the soil [sup.15]N enrichment, as observed in Forest and NTS areas, where plant matter deposition was significant and biogenic aggregates contained a high TOC content (Table 2). On the other hand, the physicogenic aggregates contained lower TOC levels (Table 2) and plant matter incorporation because they arc essentially formed by wetting and drying cycles.

Jouquet et al. (2009) found no differences in [sup.15]N levels between biogenic and physicogenic aggregates, and did not consider [sup.15]N a good indicator of aggregate type. However, considering their findings along with the results obtained in the present study, it is suggested that a combination of [sup.15]N and [sup.13]C assessment is necessary to indicate differences between the aggregate types in Forest and NTS areas.

The highest C: N ratio found in Pasture, at both depths (Table 3), was mainly related to the type of material deposited by the root system of C4 plants, in addition to the high TOC levels in this area (Table 2). No differences were found between the aggregates, probably because of the chemical stability of SOM in the Pasture area as well as the high C levels in C-HUM (Fig. 3), MOC (Fig. 2), F3 and F4 fractions (Table 4). As a result, Pasture conditions were unfavourable to the activity of soil macrofauna, mainly due to the low lability of both aggregate types.

The lowest C: N ratio was observed in the Forest area (physicogenic aggregates at 0-5 cm and both aggregate types at 5-10 cm) (Table 3). The result is related to the high N levels (especially at 5-10 cm) and low TOC levels in this area. Although the Pasture area also showed high N levels, it exhibited a higher C:N ratio than Forest (Table 3). The Forest area also received greater plant incorporation, consisting exclusively of [C.sub.3] plants. The NTS and CTS areas did not differ from each other in relation to C: N ratio, likely because of crop rotation in CTS and soybean-ryegrass succession in NTS.

The C: N ratio was higher in biogenic than physicogenic aggregates from Forest (at 0-5 cm and 5-10 cm) and NTS (at 5-10 cm), possibly because of their high TOC content (Table 2) and nutrient use (especially N) by soil macrofauna. This was mainly observed in Forest, which showed the highest occurrence of worm castings in the biogenic aggregates.

SOM fractions in the aggregates

Analysis of humic substances showed that the highest C-HUM (Fig. 3a), C-FAH (Fig. 3a, h) and C-FAF (Fig. 3a, b) levels were found in the Pasture area, probably because of the high TOC levels of both aggregate types (Table 2).

The physicogenic aggregates from the CTS area showed the lowest C-HUM (Fig. 3a, b), C-FAH (Fig. 3a) and C-FAF (Fig. 3b) levels. This result demonstrates the negative effect of ploughing and harrowing, which disrupt the aggregates, individualising particles and exposing the organic matter that was mechanically protected, inside the aggregates, against mineralisation. This process decreases carbon content in humic substances and reduces TOC (Table 2).

Biogenic aggregates from NTS exhibited lower C-HUM (Fig. 3a, b), C-FAH (Fig. 3a) and C-FAF (Fig. 3a) levels than those from Forest and Pasture areas. Physicogenic aggregates in NTS also showed lower C-HUM (Fig. 3a, b) content than those from Forest and Pasture areas and lower C-FAH and C-FAF (Fig. 3a, b) than those from Pasture, but they did not differ with respect to C-FAH and C-FAF from those from Forest. These results indicate that the length of time that the land had been managed with NTS was not sufficient to restore C-HUM, the most stable SOM fraction (Stevenson 1994), to the levels found in areas without direct human influence (Forest and Pasture). The C-FAH and C-FAF content in biogenic and physicogenic aggregates at 5-10 cm was similar between NTS and Forest areas, indicating that soil cover and the absence of soil turnover increase C levels in these fractions and at deeper layers (Fig. 3b). It means that C-FAH and C-FAF become more stratified in NTS than in Forest areas, corroborating Sa and Lai (2009) and Loss et al. (2012, 2013).

Biogenic aggregates were more suitable for the formation of humic substances. At 0-5 cm, they contained higher amounts of C-HUM, C-FAH and C-FAF than physicogenic aggregates in all the areas, and the same was observed in the Forest area at 5-10 cm. Therefore, the increase in humic substance content was also related to the activity of soil organisms (macro and microfauna) and the root system, which account for the formation of biogenic aggregates (Bronick and Lai 2005; Brussaard et al. 2007). In addition, humic substance formation is more intense on the soil surface because of plant matter incorporation and absence of soil turnover (Loss et al. 2013).

In relation to SOM particle-size fractions, POC levels were higher in biogenic and physicogenic aggregates from Pasture and Forest areas at 0-5 cm and Forest at 5-10cm (Fig. 2). Therefore, the length of time since NTS establishment was not sufficient to restore POC (the most labile SOM fraction) to the same levels recorded in areas without human interference (Forest and Pasture). Biogenic and physicogenic aggregates from the NTS area showed higher POC levels at 5-10 cm than physicogenic aggregates from the CTS area at 0-5 cm. This indicates that plant cover maintenance and the absence of soil turnover increases POC at depth (5-10 cm), and explains the greater POC stratification in the NTS than the Pasture area, as observed by Sa and Lai (2009) and Loss et al. (2012, 2013). Compared with the NTS area at 0-5 cm, the negative effects of CTS on POC are evident, indicating that ploughing and harrowing compromise POC protection inside soil aggregates.

Biogenic aggregates showed higher POC levels in all of the land-use systems and depths evaluated, demonstrating that they contain a higher content of labile material than physicogenic aggregates. Moreover, the soil fauna and root system promote POC incorporation and maintenance in biogenic aggregates. This pattern is directly related to the high TOC levels in this type of aggregate (Table 2), as well as the high soil fertility (Tables 2 and 3) and C content in humic substances (Fig. 3a, b).

In a study on three areas covered with native vegetation in the French Alps, Cecillon et al. (2010) found that POC content in biogenic aggregates was numerically (but not statistically) higher than in physicogenic aggregates, indicating the importance of biogenic processes for soil maintenance and equilibrium. Pulleman et al. (2005) also found higher POC levels in biogenic than physicogenic aggregates from areas under CTS and pasture; they reported that both aggregate types from CTS exhibited lower POC levels than those from pasture. In the present study, similar results were found in physicogenic aggregates at 0-5 cm (Fig. 2).

The high POC content in biogenic (macro) aggregates (Fig. 2) can promote the formation of microaggregates and, as a consequence, the stabilisation of recently incorporated C, as reported by Pulleman and Marinissen (2004) and Six et al. (2004). Biogenic macroaggregates seem to be important subunits for microaggrcgate formation and, consequently, for SOM stabilisation and nutrient storage, which is slowly depleted during soil degradation (Brussaard et al. 2007).

The highest MOC levels were found in the Pasture area, at both depths and for the different aggregate types, indicating that this system promotes the physical stability of SOM. This occurs because the intense activity of grass root systems favours the approximation of unitary soil particles, cements them and releases exudates. The high C: N ratio of grasses (Table 3) also contributes to SOM maintenance in the form of MOC.

The CTS area showed the lowest MOC levels at both soil depths studied, matching the lowest C-FIUM values in this area (Fig. 3a, h). This indicates that CTS was unsuitable for SOM stabilisation, because it decreased the levels of the most stable SOM fractions. MOC levels were similar between physicogenic aggregates from NTS and Forest areas at 0-10 cm and biogenic aggregates of these areas at 5-10 cm (Fig. 2). In the surface layer (0-5 cm), MOC levels in the biogenic aggregates were lower in the NTS area than in Forest and Pasture areas (Fig. 2). Because the physical stability in the NTS area is similar to that of Forest, this management system favours the increase or maintenance of SOM levels. An opposite pattern was observed in the CTS area.

The highest MOC levels were observed in biogenic aggregates at both depths. In the 0-5 cm layer, MOC levels in the two aggregate types were similar in NTS. As observed for POC in biogenic aggregates, the highest TOC (Table 2) and CHUM (Fig. 3a, b) levels in biogenic aggregates also increased MOC.

In relation to C content in the oxidisablc fractions, the CTS area showed lower levels of FI (0-10 cm), F3 (0-10 cm) and F4 (0-5 cm) fractions in physicogenic aggregates than the other areas evaluated (Table 4). These results suggest that the CTS system promotes rapid SOM mineralisation, producing lower amounts of labile organic matter (FI fraction) and, as a consequence, lower C levels in the most recalcitrant fractions (F3 and F4). With regard to F2 fraction, soil turnover and organic matter incorporation in the deepest layers produced similar C content among CTS, NTS and Forest areas at 0-5 cm and among CTS, NTS and Pasture areas at 5-10 cm.

The highest content of the most labile fraction (FI) was detected in the Pasture area, particularly in physicogenic aggregates at 0-10cm and biogenic aggregates at 0-5 cm. The F2 fraction was predominantly found in Forest and was equally distributed into physicogenic and biogenic aggregates at 0-5 cm. Pasture also showed the highest content of the F3 fraction in biogenic and physicogenic aggregates at 0-10 cm. At 0-5 cm, the levels of the F3 fraction were similar in biogenic aggregates from Forest and Pasture. The higher levels of oxidisable C in Pasture may result from the rapid growth (shoot and root) and plant mass increase of broadleaf carpetgrass, which was associated with high TOC levels (Table 2).

Compared with Forest and Pasture, the NTS area showed lower contents of F1, F2, F3 and F4 fractions in biogenic aggregates at 0-5 cm, of F1 in physicogenic aggregates at 5-10 cm, and of F2 and F3 in physicogenic aggregates at 0-5 cm. In general, the constant plant incorporation in areas of low human interference increases C levels in the oxidisable fractions of soil surface (0-5 cm). However, as indicated by C content in humic fractions (Fig. 3a, b) and particle-size fractions (Fig. 2), the oxidisable C levels at 5-10 cm were similar between Forest and Pasture areas (Table 4). This pattern indicates that soil cover maintenance in NTS favours C stratification, as reported in earlier studies (Sa and Lai 2009; Loss et al. 2012, 2013).

When differences between oxidisable C levels were detected, they were higher in biogenic aggregates (Table 4). This result shows that these aggregates are more sensitive to the management system adopted, that is, they respond directly to an increase in plant mass associated to an increase or decrease in SOM, which depends on plant-matter incorporation. Marinissen et al. (1998) evaluated the effects of microbial activity on SOM, concluding that the formation of biogenic aggregates from earthworm activity cannot be neglected during the study of organic matter dynamics in biologically active systems.

Among the systems evaluated, the Forest area deserves special attention because, compared with the Pasture and NTS areas, it exhibited higher C content in the biogenic aggregates, in the F4 fraction at 0-10cm, and in FI and F2 at 5-10 cm. This result can be explained by the activity of soil macrofauna, given that the biogenic aggregates of this area exhibited higher content of worm castings than observed in the other systems evaluated.

In general, variations in soil fertility and levels of TOC and N (Table 2), [sup.15]N and [sup.13]C (Table 3), and C in the physical and chemical SOM fractions (Figs 2 and 3, Table 4) were higher in biogenic than in physicogenic aggregates. According to Jouquet et al. (2008), this pattern is directly related to the feeding habits of earthworms (especially in Forest) and root activity (especially in Pasture). Since changes in land-use system (increase or decrease in plant material incorporation) affect soil fauna (earthworms) and the development of root systems, biogenic aggregates can be considered a reliable indicator of soil quality.

Conclusions

The CTS area, managed with ploughing and harrowing for 56 years, was unfavourable to the formation of biogenic aggregates. It promoted higher SOM mineralisation, thereby reducing the levels of TOC, N, C-HUM, POC, MOC and C in F1 and F4 fractions of physicogenic aggregates. The absence of legumes, and the long-term and constant use of N-based fertilisers, resulted in soil [sup.15]N enrichment in the CTS area (at 0-5 cm).

Successive soybean-ryegrass cropping in the NTS area for 15 years favoured aggregate formation, showing distribution of biogenic and physicogenic aggregates similar to that observed in the Forest area. Furthermore, NTS recovered C content in the physical and chemical SOM fractions (at 5-10 cm) of the biogenic and physicogenic aggregates, equalling the levels found in Forest and Pasture areas. Soil N input by chemical fertilisation, [N.sub.2] fixation through the soybean root system, and legume crop residues reduced [sup.15]N levels in the NTS area.

In the Forest area, variations between the chemical attributes (Ca, Mg, N, C-HUM and F2 and F4 fraction, at 5-10cm) of the aggregate types were more apparent than in the other systems. The low [sup.15]N values in the two layers and aggregates types of the Forest area were attributed to the absence of N-based fertilisation and low rate of biological [N.sub.2] fixation.

The Pasture system increased biogenic aggregate mass and favoured the formation of stable SOM, with high C-HUM and MOC levels in the aggregates. The high [sup.15]N levels in the Pasture area (compared with NTS and Forest areas) resulted from the absence of legumes and the constant deposition of cattle manure. It can also be associated with a more stable organic matter (with a higher degree of humification).

The biogenic aggregates were more efficient in increasing soil fertility and C content of SOM fractions than were physicogenic aggregates. We conclude, therefore, that biogenic, rather than physicogenic aggregates, can be reliable indicators of soil quality.

Information regarding topsoil aggregation status is a simple indicator of soil quality. Soil aggregates are sensitive to impacts caused by natural and anthropogenic processes, which affect the pathways for aggregate formation. This was confirmed by the absence of biogenic aggregates in the CTS area.

http://dx.doi.org/10.1071/SR14045

Acknowledgements

The study is part of the postdoctoral research of Arcangelo Loss, conducted at the Soil Science Department, at the Rural Federal University of Rio de Janeiro. The authors thank CNPq for providing Arcangelo Loss with a PhD scholarship; to Agrisus Foundation for sponsoring the research project (PA 893/11).

Received 4 October 2013, accepted 21 May 2014, published online 15 September 2014

References

Amezketa E (1999) Soil aggregate stability: A review. Journal of Sustainable Agriculture 14, 83-151. doi:10.1300/J064v14n02_08

Anghinoni I, Moraes A, Carvalho PCF, Souza F.D, Conte O, Lang CR (2011) Beneficios da integraqao lavoura-pecuaria sobre a fertilidade do solo cm sistema plantio direto. In 'Fertilidade do solo e nutriqao de plantas no sistema plantio direto'. (Eds AF Fonseca, EF Caires, G Barth) pp. 1-31. (AEACG/Inpag: Ponta Grossa, Brazil)

Barthes BG, Roose E (2002) Aggregate stability as an indicator of soil susceptibility to runoff and erosion; validation at several levels. Catena 47, 133-149. doi: 10.1016/S0341-8162(01)00180-1

Batista I, Correia MEF, Pereira MG, Bieluczyk W, Schiavo JA, Mello NA (2013) Caracterizayao dos agregados em solos sob cultivo no cerrado, MS. Semina: Ciencias Agrarias 33, 1535-1548.

Benites VM, Madari B, Machado PLOA (2003) Extrafao e fracionamento quantitative de substancias humicas do solo: um procedimento simplificado de baixo custo. Comunicado Tecnico 16, 2003. Embrapa Solos, Rio de Janeiro.

Bossuyt FI, Six J, Hendrix PF (2005) Protection of soil carbon by microaggregates within earthworm casts. Soil Biology & Biochemistry 37, 251-258. doi: 10.1016/j.soilbio.2004.07.035

Briedis C, Sa JCM, Caires EF, Navarro JF, Inagaki TM, Boer A, Neto CQ, Ferreira AO, Canalli LB, Santos JB (2012) Soil organic matter pools and carbon-protection mechanisms in aggregate classes influenced by surface liming in a no-till system. Geoderma 170, 80-88. doi: 10.1016/j.geoderma.2011.10.011

Bronick CJ, Lai R (2005) Soil structure and management: a review. Geoderma 124, 3-22. doi:10.1016/j.geoderma.2004.03.005

Brussaard L, Pulleman MM, Oue'draogo E, Mando A, Six J (2007) Soil fauna and soil function in the fabric of the food web. Pedobiologia 50, 447M62. doi: 10.1016/j.pedobi.2006.10.007

Bullock P, FcderoffN, Jongerius A, Stoops G, Tursina T (1985) 'Handbook for soil thin section description.' (Waine Research Publications: Albrighton, UK)

Cambardella CA, Elliott ET (1992) Particulate soil organic-matter changes across a grassland cultivation sequence. Soil Science Society of America Journal 56, 777-783. doi: 10.2136/sssaj 1992.03615995005 600030017x

Cecillon L, Mello NA, Danieli S, Brun J (2010) Soil macroaggregate dynamics in a mountain spatial climate gradient. Biogeochemistry 97, 31-413. doi: 10.1007/s 10533-009-9341-9

Chan KY, Bowman A, Oates A (2001) Oxidizidable organic carbon fractions and soil quality changes in an oxic paleustalf under different pasture ley. Soil Science 166, 61-67. doi: 10.1097/00010694-200101000-00009

Costa C Piccolo MC, Siqueira Neto M, Bemoux M (2012) Carbono em agregados do solo sob vegetaqao nativa, pastagem e sistemas agricolas no bioma Cerrado. Revista Brasileira de Ciencia do Solo 36, 1311 1322. doi: 10.1590/S0100-06832012000400025

Cruscio LCAC, Borghi E (2007) Consorcio de milho com braquiaria: produyao de forragem e palhada para o plantio direto. Revista Plantio Direto 100, 10-14.

Embrapa (1997) 'Manual de Metodos de analise de solo.' 2nd edn (Empresa Brasileira de Pesquisa Agropecuaria: Rio de Janeiro)

Embrapa (2006) 'Sistema Brasileiro de Classificafao de Solos.' 2nd edn. Brasilia: Embrapa Produqao de informafao, Embrapa Solos. (Empresa Brasileira de Pesquisa Agropecuaria: Rio de Janeiro)

Fiuza SS, Kusdra JF, Furtado DT (2011) Caracterizaqao quimica e atividade microbiana de coprolitos de Chibui bari (Oligochaeta) e do solo adjacente. Revista Brasileira de Ciencia do Solo 35, 723-728. doi: 10.1590/S0100-06832011000300007

Freire LR, Campos DVB, Anjos LHC, Zonta E, Pereira MG, Bloise RM, Moreira GNC, Eira PA (2013) Analise quimica de amostras de terra. In 'Manual de calagem e adubagao do Estadodo Rio de Janeiro'. I edn (Eds Freire LR, Balieiro FC, Zonta E, Anjos LHC, Pereira MG, Lima E, Guerra JGM, Ferreira MBC, Leal MAA, Campos DVB, Polidoro JC) pp. 87-100. (Embrapa/Universidade Rural: Seropedica, Brazil)

Jouquet P, Podwojewski P, Bottinelli N, Mathieu J, Ricoy M, Orange D, Tran TD, Valentin C (2008) Above-ground earthworm casts affect water runoff and soil erosion in Northern Vietnam. Catena 74, 13-21. doi: 10.1016/j.catena.2007.12.006

Jouquet P, Zangerle A, Rumpel C, Brunet D, Bottinelli N, Tran Due T (2009) Relevance and limitations of biogenic and physicogenic classification: a comparison of approaches for differentiating the origin of soil aggregates. European Journal of Soil Science 60, 1117 1125. doi: 10.1111/j. 1365-2389.2009.01168.x

Jouquet P, Phuong NT, Hanh NH, Henry-des-Tureaux T, ChevallierT, Tran Due T (2011) Laboratory investigation of organic matter mineralization and nutrient leaching from earthworm casts produced by Amynthas khami. Applied Soil Ecology 47, 24-30.

Karamanos RE, Rennie DA (1978) Nitrogen isotope fractionation during ammonium exchange reactions with soil clay. Canadian Journal of Soil Science 58, 53-60. doi:10.4141/cjss78-005

Karamanos RE, Rennie DA (1981) The isotope composition of residual fertilizer nitrogen in soil columns. Soil Science Society of America Journal 45, 316-321. doi: 10.2136/sssaj 1981.03615995004500020018x

Langenbach T, Aquino AM, Brunninger B (2002) Effects of earthworm Pontoscolex corethrurus on distribution of acaricida dicofol in a Podzolic soil. Pesquisa Agropecuaria Brasileira 37, 1663-1668. doi: 10.1590/S0100-204X2002001100019

Lavelle P, Rangel P, Kanyonyo J (1983). Mufusproduction by two species of tropical earthworms: Millsonia lamtoiana (Megascolecidae) and Pontoscolex corethrurus (Glossoscolecidae). In 'New trends in soil biology'. (Ed. P Lebrum) pp. 405-410. (Dieu Brichart Press: Louvain-la-Neuve, Belgium)

Loss A, Pereira MG, Schultz N, Anjos LHC, Silva EMR (2009) Carbono e fragoes granulometricas da materia organica do solo sob sistemas de produfao organica. Ciencia Rural 39, 1067 1072. doi: 10.1590/S0103-84782009005000036

Loss A, Pereira MG, Anjos LHC, Giacomo SG, Perin A (2011) Agregagao, carbono e nitrogenio em agregados do solo sob plantio direto com integragao lavoura-pecuaria. Pesquisa Agropecuaria Brasileira 46, 1269-1276. doi: 10.1590/S0100-204X2011001000022

Loss A, Pereira MG, Perin A, Coutinho FS, Anjos LHC (2012) Particulate organic matter in soil under different management systems in the Brazilian Cerrado. Soil Research 50, 685-693. doi: 10.1071/SR12196

Loss A, Pereira MG, Perin A, Beutler SJ, Anjos LHC (2013) Oxidizable carbon and humic substances in rotation systems with brachiaria/livestock and pearl millet/no livestock in the Brazilian Cerrado. Spanish Journal of Agricultural Research 11, 217-231. doi: 10.5424/sjar/2013111-3416

Loss A, Pereira MG, Costa EM, Beutler SJ (2014) Fragoes granulometricas e oxidaveis da materia organica sob diferentcs sistemas de uso do solo no Parana. Bioscience Journal 30, 43-54.

Marinissen JCY, Winding A, Hendriksen NB, Pulleman MM (1998) Organic matter dynamics are affected by microbial activity in biogenic aggregates. In 'Proceedings 16th World Congress of Soil Science'. Montpellier, France, 20-26 August. (International Union of Soil Sciences) (CD-ROM) Available at: www.iuss.org/index.php? option=com_content&view=article&id=10&ltemid=12

Martin A, Mariotti A, Balesdent J, Lavelle P, Vuattoux R (1990) Estimate of organic matter turnover rate in a savanna soil by 13C natural abundance measurements. Soil Biology & Biochemistry 22, 517-523. doi: 10.1016/0038-0717(90)90188-6

Natelhoffer KJ, Fry B (1988) Controls on natural nitrogen-15 and carbon-13 abundances in forest soil organic matter. Soil Science Society of America Journal 52, 1633-1640. doi: 10.2136/sssaj 1988.03615995005200060024x

O'Brien BJ, Stout JD (1978) Movement and turnover of soil organic matter as indicated by carbon isotope measurements. Soil Biology & Biochemistry 10, 309-317. doi: 10.1016/0038-0717(78)90028-7

Peres G, Cluzeau D, Curmi P, Hallaire V (1998) Earthworm activity and soil structure changes due to organic enrichments in vineyard systems. Biology and Fertility of Soils 27, 417-424. doi: 10.1007/s003740050452

Pulleman M (2002) Interactions between soil organic matter dynamics and soil structure as affected by farm management. PhD Thesis, Wageningen University, The Netherlands.

Pulleman MM, Marinissen JCY (2004) Physical protection of mineralizable C in aggregates from long-term pasture and arable soil. Geoderma 120, 273-282. doi: 10.1016/j.geoderma.2003.09.009

Pulleman MM, Six J, van Breemen N, Jongmans AG (2005) Soil organic matter distribution and microaggregate characteristics as affected by agricultural management and earthworm activity. European Journal of Soil Science 56, 453-467.

Rillig MC, Mummey DL (2006) Mycorrhizas and soil structure. New Phytologist 171, 41-53. doi: 10.1 111/j. 1469-8137.2006.01750.x

Sa JCM, Lai R (2009) Stratification ratio of soil organic matter pools as an indicator of carbon sequestration in a tillage chronosequence on a Brazilian Oxisol. Soil & Tillage Research 103, 46-56. doi: 10.1016/j.still.2008.09.003

SAEG (2007) 'Sistema para Analises Estatisticas, Versao 9.1: Fundacao Arthur Bernardes.' (Universidade Federal de Vicosa: Vicosa--MG, Brazil)

Schrader S, Zhang H (1997) Earthworm casting: stabilization or destabilization. Soil Biology & Biochemistry 29, 469 475. doi: 10.1016/S0038-0717(96)00103-4

Schumacher BA (2002) 'Methods for the determination of total organic carbon (toe) in soils and sediments.' (Ecological Risk Assessment Support Center, US Office of Research and Development: Las Vegas, NV, USA)

Silva Neto LF, Silva IF, Inda AV, Nascimento PC, Bortolon L (2010) Atributos fisicos e qulmicos de agregados pedogenicos e de coprolitos de minhocas em diferentes classes de solos da Paraiba. Ciencia e Agrotecnologia 34. 1365-1371.

Six J, Elliott ET, Paustian K (2000) Soil macroaggregate turnover and microaggregate formation: a mechanism for C sequestration under notillage agriculture. Soil Biology & Biochemistry 32, 2099-2103. doi: 10.1016/S0038-0717(00)00179-6

Six J, Bossuyt H, Degryze S, Denef K (2004) A history of research on the link between (micro) agregates, soil biota, and soil organic matter dynamics. Soil & Tillage Research 79, 7-31. doi: 10.1016/j.still.2004. 03.008

Soil Survey Staff (2010) 'Keys to Soil Taxonomy.' 11th edn. USDA Natural Resources Conservation Service. (USDA: Washington, DC)

Stevenson FJ (1994) 'Humus chemistry: Genesis, composition, reactions.' (John Wiley and Sons: New York)

Swift RS (1996) Organic matter characterization. In 'Methods of soil analysis'. (Eds Sparks DL, Page AL, Helmke PA, Loeppert RH, Soltanpour PN, Tabatabai MA, Johnston CT, Sumner ME) pp. 1011-1020. (Soil Science Society of America, American Society of Agronomy: Madison, WI, USA)

Szpak P (2014) Complexities of nitrogen isotope biogeochemistry in plant-soil systems: implications for the study of ancient agricultural and animal management practices. Frontiers in Plant Science 5, 1-19.

Teske R, Almeida JA, Hoffer A, Lunardi Neto A (2013) Caracterizcao mineralogica dos solos derivados de rochas efusivas no Planalto Sul de Santa Catarina, Brazil. Revista de Ciencias Agroveterindrias 12, 187-198.

Tisdall JM, Oades JM (1982) Organic matter and water-stable aggregates in soils. Journal of Soil Science 33, 141 163. doi: 10.1111/j. 1365-2389.1982.tb01755.x

Tivet F, Sa JCM, Lai R, Briedis C, Borszowskei PR, Santos JB, Farias A, Hartman DC, Nadolny M Bouzinac S, Seguy L (2013) 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. doi: 10.1016/j.still.2012.09.004

Velasquez E, Pclosi C, Brunet D, Grimaldi M, Martins M, Rendeiro AC, Barrios E, Lavelle P (2007) This ped is my ped: Visual separation and near infrared spectra allow determination of the origins of soil macroaggregates. Pedohiologia 51, 75-87. doi:10.1016/j.pedobi.2007. 01.002

Vezzani FM, Mielniczuk J (2011) Agregafao e estoque de carbono em argissolo submetido a diferentes praticas de manejo agricola. Revista Brasileira de Ciencia do Solo 35, 213 223. doi:10.1590/S0100-06832011000100020

Vitorello VA, Cerri CC, Anderson F, Feller C, Victoria RL (1989) Organic matter and natural carbon-13 distribution in forested and cultivated oxisols. Soil Science Society of America Journal 53, 773-778. doi: 10.2136/sssaj 1989.03615995005300030024x

Yeomans JC, Brcmner JM (1988) A rapid and precise method for routine determination of organic carbon in soil. Communications in Soil Science and Plant Analysis 19, 1467 1476. doi: 10.1080/00103628809368027

Yoneyama T (1996) Characterization of natural lsN abundance of soils. In 'Mass spectrometry of soils'. (Eds TW Boutton, S Yamsahi) pp. 225-246. (Marcel Dekker: New York)

Arcangelo Loss (A,C), Marcos Gervasio Pereira (B), Elias Mendes Costa (B), and Sidinei Julio Beutler (B)

(A) Center of Agricultural Sciences, Department of Rural Engineering, Federal University of Santa Catarina, Florianopolis, SC 88034-000, Brazil.

(B) Department of Soils, Institute of Agronomy, Rural Federal University of Rio de Janeiro, Seropedica, RJ 2389-000, Brazil.

(C) Corresponding author. Email: arcangeloloss@yahoo.com.br; arcangelo.loss@ufsc.br

Table 1. Relative mass contribution (%) of biogenic and physicogcnic
aggregates in soil under different land-use systems in Marmeleiro,
Parana State, Brazil

Measurement corresponds to 100g of soil aggregates before separation
between biogenic and physicogenic. CTS, Conventional tillage system;
NTS, no-till system; CV, coefficient of variation. Within columns,
means (four replications per field) followed by the same uppercase
letter are not significantly different between land-use systems for
each aggregate type, and within rows, means followed by the same
lowercase letter are not significantly different between aggregate
types in the same system (l.s.d. at P=0.05)

Land-use systems   %Biogenic    %Physicogenic
                   aggregates    aggregates

                   0-5 cm

CTS                  0.00Cb       100.00Aa
NTS                  9.55Bb        90.45Ba
Forest              12.88Bb        87.12Ba
Pasture             27.90Ab        72.10Ca
CV (%)              22.3           19.2

                  5-10 cm

CTS                  0.00Cb       100.00Aa
NTS                  8.34Bb        91.66Ba
Forest               9.74Bb        90.26Ba
Pasture             23.78Ab        76.22Ca
CV (%)              17.2           14.2

Table 2. Chemical attributes of biogenic (Bio) and physicogenic (Phy)
aggregates in soil under different land-use systems in Marmelerio
Parana State, Brazil

CTS, Conventional tillage system; NTS, no-till system; CV,
coefficient of variation; TOC, total organic carbon. Within columns,
means (four replications per field) followed by the same uppercase
letter are not significantly different between land-use systems for
each aggregate type, and within rows and parameters. means followed
by the same lowercase letter are not significantly different between
aggregate types in the same system (l.s.d. at P = 0.05)

Land-use   pH ([H.sub.2]0)     Ca ([cmol.sub.c]    Mg ([cmol.sub.c]
systems                        [kg.sup.-1])        [kg.sup.-1])

             Bio       Phy       Bio       Phy       Bio       Phy

                               0-5 cm

CTS          --      5.69BC      --       5.80C      --      3.20C
NTS        6.37Aa    6.25Aa     9.43Ba    8.27Ba   4.73Ba    5.16Aba
Forest     6.15Aa    6.08ABa   13.00Aa   10.33Ab   6.43Aa    4.80Bb
Pasture    5.32Ba    5.29Ca     5.86Ca    4.50Cb   6.56Aa    5.77Aa
CV (%)     4.7       4.1        8.3       9.6      9.9       8.9

                              5-10 cm

CTS          --      6.15A       --       5.76B       --      3.10B
NTS        5.91Aa    5.69Ba     7.33Ba    7.07ABa    4.96Aa   4.46Aa
Forest     5.83Aa    5.68Ba    12.23Aa    8.06Ab     5.40Aa   4.30Ab
Pasture    4.97Ba    4.83Ca     1.90Ca    1.73Ca     3.03Ba   2.90Ba
CV (%)     4.7       4.4       17.1      18.3       11.4      9.8

Land-use   K. ([cmol.sub.c]    Na ([cmol.sub.c]    Al ([cmol.sub.c]
systems    [kg.sup.-1])        [kg.sup.-1])        [kg.sup.-1])

             Bio       Phy       Bio       Phy

                               0-5 cm

CTS          --       0.76C       --        0.04B       --      0.01B
NTS         0.77Ca    0.83BCa    0.02Ba     0.02Ba    0.01Aa    0.01Ba
Forest      1.67Ba    1.29Ba     0.05Aa     0.02Bb    0.01Aa    0.01Ba
Pasture     2.85Aa    2.48Aa     0.04ABb    0.08Aa    0.01Ab    0.05Aa
CV (%)     13.2      20.2       26.1       19.9      30.0      42.0

                              5-10 cm

CTS          --       0.64B       --      0.02B       --      0.01B
NTS         0.70Ba    0.77Ba    0.04Aa    0.01Bb    0.01Ba    0.01Ba
Forest      0.79Ba    0.64Ba    0.03Aa    0.01Bb    0.01Ba    0.01Ba
Pasture     1.87Aa    1.80Aa    0.04Aa    0.05Aa    0.42Aa    0.51Aa
CV (%)     11.4      10.6      25.9      34.4      43.0      30.3

Land-use   H + Al ([cmol.sub.c]   P (mg [L.sup.1])   N (g [kg.sup.-1])
systems    [kg.sup.-1])

                               0-5 cm

CTS           --      6.21B        --      6.37A        --    1.95D
NTS         4.73Bb    5.61Ba    13.92Aa    8.23Ab    3.14Ba   2.78Ca
Forest      5.77Ba    4.84Bb     5.04Ba    2.91Bb    5.48Aa   3.67Bb
Pasture    11.93Aa   11.00Aa    14.16Aa    6.74Ab    5.53Aa   4.35Aa
CV (%)     13.3      13.7       15.4      22.6       9.9      8.2

                              5-10 cm

CTS           --      6.16B       --       6.44A        --      1.96B
NTS         5.72Ba    6.16Ba   12.32Aa     5.40Ab     2.64Ba    2.34Ba
Forest      6.71Ba    6.60Ba    2.55Ba     1.74Bb     3.87Aa    3.14Aa
Pasture    11.82Aa   11.16Aa    2.24Ba     1.78Bb     2.90Ba    2.57ABa
CV (%)     11.4       8.9      12.8       25.9       17.1      13.2

Land-use   TOC (g [kg.sup.-1])
systems

            0-5 cm

CTS          --      18.44C
NTS        32.07Ca   25.38Bb
Forest     46.24Ba   29.72Bb
Pasture    59.29Aa   47.24Ab
CV (%)      5.4       7.7

           5-10 cm

CTS          --      18.28C
NTS        28.82Ca   22.32BCb
Forest     30.35Ba   26.36ABb
Pasture    33.25Aa   29.45Ab
CV (%)      7.6      11.5

Table 3. Natural [sup.15]N and [sup.l3]C abundance in biogenic (Bio)
and physicogenic (Phy) aggregates in different land-use systems in
Marmeleiro, Parana State, Brazil

CTS, Conventional tillage system; NTS, no-till system; CV,
Coefficient of variation. Within columns, means (four replications
per field) followed by the same uppercase letter arc not
significantly different between land-use systems for each aggregate
type, and within rows and parameters, means followed by the same
lowercase letter are not significantly different between aggregate
types in the same system (l.s.d. at P = 0.05)

Land-use   [sup.15]N%o        [sup.l3]C%o           C: N
system
            Bio       Phy       Bio        Phy        Bio       Phy

0-5 cm

CTS          --     8.03A        --       22.81 B      --      9.42B
NTS        6.09Bb   6.53Ca     24.71Bb    24.21Ca    9.38Ba    9.12Ba
Forest     4.27Cb   5.35Da     26.74Cb   -25.97Da    8.82Ba    8.07Cb
Pasture    7.44Aa   7.52Ba    -16.12Aa    16.41Ab   10.74Aa   10.86Aa
CV (%)     7.4      3.6         2.1        1.9       5.2       3.5

5-10 cm

CTS          --     7.94AB       --      -22.81B      --       9.25BC
NTS        7.04Bb   7.40BCa   -24.38Bb   -23.68Ca   10.15Ba    9.50Bb
Forest     5.92Cb   6.88Ca    -25.98Cb   -25.29Da    8.87Ca    8.44Cb
Pasture    8.27Aa   8.28Aa     17.89Aa    18.12Ab   11.50Aa   11.47Aa
CV (%)     7.0      4.2         2.5        1.8       4.8       4.9

Table 4. Carbon content (g [kg.sup.-1]) in the oxidisable fractions
of biogenic (Bio) and physicogenic (Phy) aggregates from different
land use systems in Marmeleiro, Parana State, Brazil

Fractions correspond to the part of organic C oxidised by
[K.sub.2][Cr.sub.2][0.sub.7] in acid solution with: F1,
[H.sub.2]S[0.sub.4] <3 mol [L.sup.-1] F2, 6-3 mol [L.sup.-1] F3, 9-6
mol [L.sup.-1] and F4, 12-9 mol [L.sup.-1] CTS, Conventional tillage
system; NTS, no-till system; CV, Coefficient of variation. Within
columns, means (four replications per field) followed by the same
uppercase letter arc not significantly different between land-use
systems for each aggregate type, and within rows and parameters,
means followed by the same lowercase letter are not significantly
different between aggregate types in the same system (l.s.d. at P =
0.05)

Land-use            F1                    F2
system
             Bio        Phy        Bio        Phy

                       0-5 cm

CTS           --       5.50C        --       5.33BC
NTS        10.16Ca     9.50Ba     9.73Ba     5.00Cb
Forest     17.27Ba     9.17Bb    11.37Aa     6.67Bb
Pasture    21.83Aa    16.67Ab    12.33Aa     9.33Ab
CV (%)      6.3        8.3        6.9       12.0

                       5-10 cm

CTS           --       5.17D        --       5.17A
NTS        10.33Ba     8.00Cb     5.33Ba     5.33Aa
Forest     12.33Aa     8.83Bb     8.00Aa     2.50Bb
Pasture    10.50Ba     9.83Aa     5.33Ba     5.83Aa
CV (%)      7.6        5.4       11.0       17.6

Land-use            F3                     F4
system
             Bio         Phy        Bio        Phy

                       0-5 cm

CTS           --       2.66D         --       1.83B
NTS         4.50Ba     5.00Ca      3.83Ca     4.23Aa
Forest     10.00Aa     6.75Bb     10.07Aa     4.33Ab
Pasture    11.67Aa    12.0Aa       5.83Ba     4.00Aa
CV (%)     13.7       12.3        10.5       13.7

                       5-10 cm

CTS           --       2.66C         --       3.17A
NTS         4.75Ba     3.83BCa     4.16Ba     3.73Aa
Forest      5.83Ba     5.33Ba      6.33Aa     3.33Ab
Pasture     9.33Aa     8.00Aa      4.66Ba     3.50Aa
CV (%)     15.0       17.4        15.1       18.3
COPYRIGHT 2014 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Loss, Arcangelo; Pereira, Marcos Gervasio; Costa, Elias Mendes; Beutler, Sidinei Julio
Publication:Soil Research
Article Type:Report
Date:Oct 1, 2014
Words:10851
Previous Article:Coastal acid sulfate soils in the Saloum River basin, Senegal.
Next Article:Soil fertility changes following conversion of grassland to oil palm.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters