INTRODUCTION

The organic carbon content is one of the major factors controlling the aggregate stability of soils [2, 24, 28]. Soil aggregation protects the organic matter (OM) and affects the rate of its turnover and the soil fertility [13, 28]. The optimum soil structure should be tolerant to different moisture conditions and should ensure continuity of pores in the soil matrix to create favorable conditions for the water and air exchange between the roots and the soil.

Soil aggregates can be considered secondary structural units consisting of the combinations of mineral particles with organic and/or inorganic binding substances [16, 23]. The components of OM bind primary mineral particles by physical and chemical bonds into the aggregates and favor aggregate stability; in particular, OM protects aggregates from slaking upon moistening [30].

Most of the conceptual models of soil aggregation assume that soils consist of the aggregates of different sizes with organic and inorganic binding substances. One of the concepts suggests that the formation of macroaggregates is the result of systematic binding of microaggregates that, in turn, are the products of gluing of ultramicroparticles (<2 µm) representing organomineral complexes [27, 40]. Other concepts postulate that the breakdown of macroaggregates (>250 µm) into microaggregates (20 to 250 µm) is a preliminary stage of the formation of microaggregates. In other words, it is assumed that microaggregates are mainly formed within macroaggregates [33, 38, 41]. According to the current state of knowledge on the models of soil aggregation, the aggregate-protected pool of OM favors stabilization of microaggregates; in turn, the latter protect the OM from microbial attacks. The incrustation of OM in the center of microaggregates are the major way of the organic carbon sequestration [33, 40]. The degree of protection of OM in the aggregates is proportional to the specific surface of clay particles and to the interaction between surface monolayers of sand and clay particles [29].

The study of water-stable soil structure is used to judge the degree of soil aggregation and the strength of aggregates. Aggregate saturation with water may have a destructive effect, because, upon rapid soil moistening, the release of air entrapped in aggregate pores creates high pressure, which breaks the aggregate. This is the way of aggregate destruction into fragments and primary particles [30]. Therefore, the quality and stability of the aggregate state of soils mainly depends on the water stability of aggregates. Water-stable structure is usually composed of macro- (>250 µm) and micro- (<250 µm) aggregates [24]. The stability of aggregates provided by adhesion depends on the aggregate size [4]. With an increase in the size of aggregates, the adhesion force decreases and, hence, the stability of aggregates becomes lower [40]. Hence, the effects favoring aggregate stability are better manifested in smaller aggregate fractions. The smaller the size of aggregates, the higher their tolerance towards these destructive forces. The presence of binding agents, such as humic substances associated with amorphous iron and aluminum compounds ensures the high stability of microaggregates.

The stability of macroaggregates is lower than the stability of microaggregates, because macroaggregates are partly stabilized by short-living binding agents, such as roots, fungal hyphae, and polysaccharides produced by microorganisms and plants. The aggregation at the macrolevel is much more sensitive to changes of the cenoses than the aggregation at the microlevel [8, 18, 24, 41].

Water stability of aggregates depends on the organic carbon content of the aggregates and directly affects the structural state and physical properties of the soil. Organic compounds ensure the stability of soil aggregates via lowering their wettability and affecting their mechanical strength [35, 44]. In particular, OM bound with phyllosilicates increases their hydrophobicity and makes these aggregates more tolerant to the soil moistening [25].

The composition of OM and the factors affecting the stability of soil microaggregates with different physical strengths have been studied in several research projects [14, 17, 27, 32, 34, 36, 37, 40, 42]. However, quantitative information on the transformation of OM in micro- and macroaggregates with due account for the particular fractions of OM isolated by the methods of physical fractionation from chernozems under different land uses is still insufficient.

In this context, the aims of our research were as follows: (1) to isolate water-stable microaggregates from the samples of typical chernozem, (2) to isolate microaggregates with different stability toward the ultrasonic treatment from the water-stable macro- and microaggregates within the 2–1 mm air-dry aggregates, and (3) to characterize the content and composition of OM in water-stable macro- and microaggregates of typical chernozem.

OBJECT AND METHODS

We studied typical chernozems under different land uses: (a) virgin steppe (in the Central Chernozemic State Biospheric Reserve Streletskaya Steppe (Kursk oblast) and (b) long-term (52 yr) permanent bare fallow without application of fertilizers (experimental field of the Kursk Research Institute of Farming (51°34′ N, 36°06′ E). The plot with permanent bare (tilled) fallow (15 × 200 m) was established in 1964. In 1998, it was modified: two-thirds of the initial area was left under permanent bare fallow, and one-third was converted into the overgrown (unmanaged) fallow [5].

The territory of studied sites is typical of the Central chernozemic region of European Russia. The climate is moderately cold with the mean annual temperature of 4.8–5.3°C. The mean annual precipitation reaches 540 mm. The soil is classified as a silt loamy typical chernozem [6], or as a Haplic Chernozem (Loamic, Pachic) [45]. It is developed from loesslike sediments with the clay (<1 μm) content of 18.4–22.9%, the Corg content of 2.6–4.8%, and pHH2O 6.8–6.9.

From each experimental area, three undisturbed soil monoliths (25 × 25 × 15 cm) were sampled. Before physical fractionation, these monoliths were broken into aggregates via their throwing from the height of 1.5 m onto the surface covered by polyethylene film. Then, under laboratory conditions, the soil samples were placed onto the flat surface, and coarse plant residues and roots were manually taken off from them. Then, the samples were air-dry under shade to exclude direct sunlight action on them. After dry sieving of the samples, the following fractions of air-dry aggregates were obtained: >10, 10–7, 7–5, 5–3, 3–2, 2–1, 1–0.5, 0.5–0.25, and <0.25 mm. Then, air-dry aggregates 2–1 mm were subjected to wet sieving according to the Savvinov method in modification by Khan [13]. Aggregate fraction 2–1 mm was chosen because of the greatest contribution of OM of this fraction to the total soil OM [8, 11]. Aggregates taken from the 1-mm sieve were presaturated with water via spraying in order to avoid the destruction of capillaries by the entrapped air and then subjected to wet sieving in standing water. Thus, we obtained water-stable aggregate fractions 2–1, 1–0.5, and 0.5–0.25 mm and water-stable microaggregates (<0.25 mm). Then, water-stable macro- and microaggregates were air dried and sieved through a 1-mm sieve. The sieved material was comminuted with a rubber pestle and homogenized. Rubber pestle was used to avoid excessive destruction of the aggregates and crushing of primary minerals.

To study the OM composition, macro- and microaggregates were subjected to the granulodensimetric fractionation.

Currently, various methods are used to study the distribution of carbon in fractions of aggregates of different sizes [21]. The results obtained depend on the choice of the method [19]. In our study, a modified method of granulodensimetric fractionation was used [2, 12]. This method allows one to divide microaggregates into two groups differentiated according to their tolerance toward the dispersing effect of ultrasound and significantly differing in the properties of their organic and organomineral components. In accordance with the concept of microaggregate stability [14, 27], the components of the soil, which were separated after a short (5–15 min) ultrasonic treatment, are included in large (unstable) microaggregates of 50–250 µm in size. The components of the soil residue after the separation of large microaggregates enter the composition of small (stable) microaggregates of 1–50 µm in size.

According to the methodological scheme [2], free (nonoccluded) OM localized in the interaggregate space of water-stable aggregates is extracted with bromoform–ethanol mixture as the free light (<1.8 g/cm3) fraction (LFfr). Then, aggregates are destroyed by ultrasonic treatment. For the physical dispersion, an ultrasonic disperser LUZD-0.5K-02-00000 PS (Kriamid, Russia) was used. Ultrasonic treatment (71 J/mL) of the soil sample (10 g + 50 mL of deionized water) continued for 1 min and was followed by centrifugation in accordance with Stokes’ Law. This procedure was repeated 15 times. Water suspension of clay particles (<1 μm) was collected and dried (t = 80°C) [15, 39]. The repeated procedure of ultrasonic dispersion of low intensity and extraction of clay particles ensured gradual destruction of microaggregates; note that the removal of each new portion of clay particles from the initial sample prevented them from further destruction under the impact of excessive cavitation. The choice of the <1 μm (clay) fraction was specified by a greater homogeneity of OM in this fraction in comparison with that in the fraction of <2 μm [20, 26].

After removal of the clay particles from the sample, the occluded OM localized within microaggregates and having the density <1.8 g/cm3 (occluded light fraction, LFocc) was isolated using the bromoform–ethanol mixture. All the separated light fractions were subdivided by sieving into two subfractions: larger and smaller than 50 μm. A study with the use of the modified method by Golchin with coauthors [31] applied to soil samples of different genesis [3] demonstrated a predominance of undecomposed and partially decomposed fragments of plant tissues in the first (>50 µm) subfraction and of smaller and more decomposed fragments of plant tissues, including fully humified tissues, in the second (<50 µm) subfraction. All isolation procedures were performed in triplicate.

Thus, the applied method separates four pools of OM: non-occluded (free) light fraction (LFfr), occluded light fraction (LFocc), OM of the clay fraction (Clay), and OM of the soil residue (Res). The application of granulodensimetric fractionation of soil samples and structural units makes it possible to separate two groups of microaggregates differing in their tolerance towards the dispersing action of ultrasound. Large (50–250 µm) microaggregates are unstable in the ultrasonic field and consist of plant tissues of different degrees of mineralization–humification (LFocc) and of clay particles (Clay). The bonds between these components are weak, so they are easily released under the impact of cavitation. Small (1–50 µm) microaggregates are stable in the ultrasonic field and remain in the Residue fraction.

The contents of carbon and nitrogen in the samples of soils, structural units, and granulodensimetric fractions were determined by the combustion catalytic oxidation method with a ТОС Analyser (Shimadzu, Japan). All measurements were performed in triplicate.

Statistical processing of the results was performed using Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA, USA) and Origin Pro 8 (Origin Lab Corporation, Northampton, MA, USA) software. The significance level P < 0.05 was chosen.

RESULTS AND DISCUSSION

The analysis of the soil structure by dry sieving showed that the aggregation degree depends on the type of land use. This is in agreement with the results obtained by Kuznetsova [9]. The content of air-dry aggregates (10–1 mm) is maximum (73.0 ± 2.3%) in the typical chernozem under steppe cenosis (Table 1). Under the 52-year-long bare fallow, their content decreases to 57.1 ± 8.0%. In the steppe chernozem, the contents of air-dry aggregates of different sizes decreased in the following sequence: (2–1)> (5–3)> (3–2)> (<0.25)> (7–5)> (0.5 –0.25)> (1–0.5)> (>10)> (10–7) mm. In the chernozem under bare fallow, a different sequence was observed: (2–1) > (<0.25) > (0.5–0.25) > (5–3) = (>10) > (1–0.5) > (3–2) > (7‒5) > (10–7) mm.

Table 1.   Distribution of air-dry aggregates and the organic carbon content in the typical chernozem under different land uses
Full size table

The analysis of the distribution and OM content of air-dry aggregates (Table 1) showed that, regardless of the type of land use, the largest mass of aggregates is in the fraction of 2–1 mm in size. However, under the bare fallow, a fivefold increase in the content of lumpy (>10 mm) fraction is observed; the content of the fraction 10–7 mm increased by more than four times. At the same time, the contents of smaller aggregate fractions (1–0.25 and 0.5–0.25 mm) also increases in comparison with that in the steppe chernozem. The content of agronomically valuable aggregates larger than 2 mm (5–3 and 3–2 mm) in the regularly tilled soil of the bare fallow decreases by more than two times in comparison with that in the steppe chernozem. These conclusions are in agreement with previously obtained data [22].

Thus, it can be stated that in the chernozem of permanent (52 yr) bare fallow, an increase in the degree of lumpiness of the soil structural units has taken place. This is accompanied by a decrease in the amount of agronomically valuable aggregates. It has been found that the 2–1 mm air-dry aggregates represent one of the most conservative elements of the soil macrostructure: regardless of the land use, their content reaches 26–27% of the total mass of aggregates.

The concentration of carbon in the air-dry aggregates of steppe chernozem, as well as in the whole soil, is high and varies in a narrow range from 4.6 to 5.1%. After 52 years of bare fallowing, it has sharply and reliably (P = 0.95) decreased to 2.4–2.7% because of the extremely low input of fresh OM into the tilled chernozem of bare fallow and the mineralization of OM in the soil.

Taking into account the maximum amount of aggregates 2–1 mm in the soil and the increased Corg content in them, it is obvious that the contribution of these aggregates to the total OM content in the soil is maximum both in the steppe and in the bare fallow; it reaches 26.6% of the total Corg content. In relation to this, we further examined air-dry aggregates of this size as the most representative aggregate fraction in typical chernozems [8, 11]. These aggregates were subjected to wet sieving.

It was found that air-dry aggregates 2–1 mm in the steppe soil are mainly (80%) represented by water-stable macroaggregates of the same size (2–1 mm) (Table 2). In the steppe chernozem, the content of water-stable aggregates decreased in dependence on their size in the following order: (2–1) > (1–0.5) > (<0.25) > (0.5–0.25) mm. After 52 years of the chernozem functioning in the regime of bare fallow, the distribution of the amount of water-stable aggregates of different sizes has radically changed: (0.5–0.25)> (1–0.5) > (<0.25) > (2–1) mm. All these differences between the contents of aggregates in the studied steppe and bare-fallow chernozems proved to be statistically significant. Thus, the long-term permanent bare fallowing of chernozem contributed to a sharp decrease in the water stability of its aggregates. This is evidenced by the almost complete disappearance of water-stable macroaggregates (2–1 mm) as a result of extreme agrotechnical loads on the soil.

Table 2.   Size distribution and carbon content of water-stable units separated from the 2–1 mm air-dry aggregates
Full size table

The content of Corg in water-stable units of different sizes in the steppe chernozem is approximately the same, except for the water-stable microaggregates <0.25 mm, in which it is significantly lower. In the chernozem under bare fallow, the Corg content in the separated water-stable units considerably decreases, except for the water-stable macroaggregates 2–1 mm, in which it is comparable with that in the steppe chernozem. However, the yield of this fraction from the chernozem of the bare fallow is extremely low.

Analytical data attest to high microaggregation of both soils as a whole and of their water-stable units of different sizes separated from the air-dry aggregates 2–1 mm. On the average, the portion of microaggregates unstable in the ultrasonic field in different macroaggregate fractions of both variants of the chernozem varies within narrow limits.

The portion of free OM in the composition of the studied water-stable units is very small and decreases with a decrease in their size. Long-term bare fallowing of the chernozem has led to a sharp decrease in the content of free organic matter both in the whole soil (by more than twelve times) and in the water-stable microaggregates (by two times). On the contrary, in the water-stable macroaggregates (2–1 mm) of the chernozem under bare fallow, the content of free OM has increased by more than five times in comparison with that in the steppe chernozem (Table 3).

Table 3.   The composition of water-stable macro- and microaggregates within the 2–1 mm air-dry aggregates in studied soils
Full size table

The materials of earlier studies [1, 3] allow us to suppose that the OM composition in the light fractions from the whole soil and from separate aggregates is fundamentally different in the chernozems of the steppe and permanent bare fallow. It is most likely that plant residues of various degrees of humification predominate in the composition of the free OM in the steppe chernozem, whereas highly condensed metal humates predominate in the free OM of the chernozem of permanent bare fallow.

Analytical data (Table 4) indicate that the maximum concentrations of carbon and nitrogen are typical of the occluded OM (LFocc) (23–34% C and 1.5–2.2% N) localized within unstable microaggregates. This conclusion is consistent with the results obtained by Golchin et al. [31]. On the average, carbon and nitrogen concentrations tend to decrease in the following order: LFocc > LFfr > Clay > Residue.

Table 4.   The contents of C and N and the C/N ratio in organic and organomineral fractions of typical chernozem and its macro- and microaggregates within the 2–1 mm air-dry aggregates
Full size table

In the steppe chernozem, the concentrations of carbon and nitrogen in the clay fraction vary within narrow ranges (7.4–8.5 and 0.8–0.9%, respectively). For the chernozem under bare fallow, these ranges are wider (4.7–6.5 and 0.5–0.7%, respectively). The lowest concentrations of carbon and nitrogen are noted for the Residue fraction.

Significant differences between the soils of different land uses are observed for the total level of carbon accumulation in the bulk soil/aggregates of different sizes. In the steppe chernozem, the organic carbon content reaches 4.8%. In the chernozem of permanent bare fallow, it decreases by more than 1.8 times (Table 5). These experimental results are consistent with the theoretical position on the limiting (maximum and minimum) level of OM accumulation in typical chernozems under conditions of steppe and long-term permanent bare fallow [7, 10].

Table 5.   The C content in the typical chernozem and its macro- and microaggregates within the 2–1 mm air-dry aggregates, %
Full size table

The loss of carbon from water-stable unites within the 2–1 mm air-dry aggregates is considerably lower than that from the bulk soil. In the water-stable microaggregates from the chernozem of bare fallow, it reaches 29% relative to the initial carbon content in the steppe chernozem. In the undisturbed water-stable macroaggregates, the carbon content in the chernozem of bare fallow decreases by only 9% relative to the initial content. Thus, we can conclude about a greater water resistance of macroaggregates in comparison with microaggregates and with the whole soil.

Long-term bare fallowing of the chernozem has resulted in a sharp decrease in the content of free OM in the bulk soil (by eleven times), whereas water-stable microaggregates have shown just a tendency for a decrease in the content of free OM. This may be related to a significant difference in the amounts of fresh plant residues entering the chernozems of these land uses. On the bare fallow plot, this input is virtually absent. In the water-stable macroaggregates of the chernozem under bare fallow, the content of free OM has increased by five times. This may be related to the nature of the light fractions in the steppe chernozem and in the chernozem under bare fallow. In the latter case, it is possible that the OM of light fractions is largely represented by the highly condensed metal humates and humus coals.

For the occluded OM of the chernozem under bare fallow, a decrease in the level of carbon accumulation (in comparison with that in the steppe chernozem) has taken place both in the whole soil and in the water-stable aggregates. On the average, a decrease in the carbon of LFocc in the whole soil and in water-stable aggregate fractions has the following pattern: whole soil (by 2.8 times) > water-stable microaggregates (by 2.1 times) > water-stable macroaggregates (by 1.4 times).

In the steppe chernozem, the organic carbon pool of the organic–clay complexes reaches 1.5 % of soil. In the chernozem under bare fallow, mineralization of the relatively more inert OM of the organic–clay complexes is 1.4 times less intense than that in the whole soil. For water-stable microaggregates, the mineralization loss of OM is somewhat lower (by 1.3 times), and for water-stable macroaggregates, the mineralization loss of carbon is minimal (by 1.1 times). The organic matter of organic–clay complexes localized within water-stable macroaggregates is more stable than that in the water-stable microaggregates and in the whole soil.

A sharp (by 1.8 times) decrease in the carbon of the Residue fraction, as well as of the whole soil, has taken place in the chernozem under bare fallow (in comparison with the steppe chernozem). This is due to mineralization of the carbon of microaggregates resistant in the ultrasonic field (inert OM) in the absence of the input of fresh organic residues into the soil under bare fallow. For water-stable macroaggregates, the loss of carbon is also large (by 1.7 times), while in water-stable microaggregates of this chernozem this loss is only about 1.1 times of the initial content.

CONCLUSIONS

It is shown that the aggregation of typical chernozem largely depends on the land use: the content of air-dry aggregates (10–1 mm) is maximum (over 70%) under the steppe cenosis. In the chernozem functioning in the regime of permanent bare fallow for 52 years, their content decreases by 1.3 times. Simultaneously, an increase in the degree of lumpiness of structural units and in the content of silt-size aggregates takes place. These phenomena are accompanied by a decrease in the number of a larger part of agronomically valuable aggregates, except for those of 2–1 mm in size.

It has been found that the content of air-dry aggregates of 2–1 mm in size in the chernozem virtually does not change even upon contrasting regime of its use (the steppe and the bare fallow).

At the same time, the water stability of air-dry aggregates of that size sharply differs in the chernozems under the steppe and under the bare fallow. Air-dry aggregates (2–1 mm) in the steppe chernozem are characterized by the high water stability: 80% of them are represented by water-stable macroaggregates of the same size (2–1 mm). After 52 years under the bare fallow, water-stable aggregates have disappeared from this fraction of air-dry aggregates almost completely.

On the average, the portion of free OM in water-stable units separated from the 2–1 mm air-dry aggregates has sharply decreased in the chernozem after 52 years under bare fallow: from two times for water-stable microaggregates to more than twelve times for the whole soil. On the contrary, in water-stable macroaggregates from this chernozem, the portion of free OM has increased by more than five times.

Analytical data attest to the high degree of microaggregation of water-stable structural units within the 2–1 mm air-dry aggregates. As a result of bare fallowing of the typical chernozem for 52 years, significant carbon losses have taken place in all the OM pools of this soil. The smallest losses have been observed for the free and occluded light fractions and for clay-bound OM in the undestroyed water-stable macroaggregates. This testifies to their greater stability in comparison with that of the water-stable microaggregates and the whole soil, as well as to a greater degree of protection of OM localized within water-stable macroaggregates. The remaining part of the water-stable macroaggregates in the chernozem under bare fallow is similar in the studied properties to the corresponding macroaggregates in the steppe chernozem.