1. Introduction
Soil organic matter (SOM) is the most important factor in soil quality due to its effect on the soil chemical, physical, and biological processes. Croplands around the world undergo changes as a result of long-term use and, in most cases, due to improper management [
1]. As a consequence, this has led to the depletion of soil resources, the development of degradation processes, the loss of soil organic carbon (SOC), and CO
2 emissions into the atmosphere [
2]. Arable land is considered the most promising for SOC accumulation for climate change mitigation among other land uses [
3]. With vast soil resources including fertile chernozems, Russia is among the top five countries with a high total additional annual SOC sequestration potential [
4]. Despite the abandonment of large amounts of arable chernozems as a result of the collapse of the Soviet Union, extensive areas of agrochernozems have continued to be used for centuries, deteriorating the soil physical and chemical properties [
5,
6,
7,
8,
9].
Agricultural use affects not only the quantitative but also the qualitative characteristics of SOM [
10,
11,
12,
13,
14]. Ploughing the soil changes the hydrothermal conditions as well as the character and rates of input of SOM, which causes the transformation of the molecular structure of humic acids (HAs) [
15]. Studies of HAs make it possible to determine the rates and limits of SOM loss under anthropogenic impacts. With the advent of high-precision methods such as
13C-NMR spectroscopy, it became possible to accurately quantify the composition and structure of SOM. The
13C-NMR spectroscopy method allows for the study of the structural and compositional features of the HA preparations, which provides a reliable assessment of the fundamental processes of humification and the composition of natural molecular-weight HAs [
16,
17].
The primary objectives of the present study were: (1) to examine the soil physical-chemical properties of long-term arable argochernozems and adjacent territory under forest windbreak; and (2) to study and compare the molecular composition of HAs isolated from different soils using 13C-NMR spectroscopy analysis.
2. Materials and Methods
This study was conducted on an arable plot involved in agricultural use for at least 50 years. The study site is located in the southern forest-steppe zone of the Republic of Bashkortostan, Russia (
Figure 1). The area of the study site is 200 hectares; the height above sea level varies from 165 m in the northwestern part to 195 m in the southeastern part. The cropland is characterized by ploughing with a turnover of the soil layer at a depth of 15–20 cm. In the year of field work and sample collection (2019), wheat (
Triticum aestivum) was cultivated, while earlier in 2018, sugar beet was grown. The soil of the study site is represented by chernozem Calcic soils according to the WRB classification [
18]. The study site is subject to the manifestation of water and wind erosion processes. In the northern and eastern parts, there are windbreaks to prevent the effects of wind erosion.
The climate of the region is warm-summer humid continental (Dfb), according to the Köppen climate classification [
19], with a mean annual temperature of 2.8–3 °C. The mean January temperature is −15 °C, with an absolute minimum of −46 °C. The mean July temperature is +19 °C with an absolute maximum of +38 °C. Winter is characterized by steady frosty weather, snowfalls, and rare thaws. Summer is warm, with occasional rainfall. The mean annual precipitation is 450–550 mm [
20].
The distribution of parent materials in the area is associated with the geological structure, the nature of the relief as well as the proximity to the floodplain terraces of the rivers. Predominant parent materials are mainly deluvial and eluvial-deluvial deposits [
21].
We collected a total of 40 soil samples, 25 of which were taken from arable land and 15 from windbreaks from the topsoil (0–20 cm) during field work after harvest (October 2019). The cropland samples were taken from erosion-prone and undisturbed areas. The soil sampling work was conducted via a stratified simple random-sampling scheme. The approach was to randomly select soil samples from predetermined areas (arable undisturbed areas, arable erosion-prone, and pristine windbreaks). We used satellite data prior to fieldwork to identify erosion-prone and undisturbed areas. Erosion areas are well-identifiable from remote sensing data and are characterized by linear furrows and lighter areas due to shallow surface flow. Additionally, one non-eroded arable soil profile was opened up to the parent material and samples were collected from each genetic horizon (Ap, A1, AB, B). The erosion sediments (A0) and samples from the A1 horizon were collected from soils under a forest windbreak. Then, all of collected soil samples were air-dried, homogenized, sieved, and involved in subsequent laboratory analyses.
Soil structure and texture measurements were performed according to the methodology of Vadyunina and Korchagina [
22]. In particular, the structural-aggregate composition (dry sieving) was determined by using meshes with sizes of 10, 7, 5, 3, 1, 0.5, and 0.25 mm. Soil aggregate stability (wet sieving) was measured with a Baksheev device; the particle size distribution was conducted according to the Kachinsky “wet sedimentation” (pipette) method, which is the Russian analogue of analysis by Bowman and Hutka [
23]. Aggregate condition was determined according to the coefficient of soil structure (Cst) (Equation (1)), where a value > 1.5 is classified as excellent; 1.5–0.67 as good, and 0.67 as unsatisfactory [
22]. Aggregate stability was calculated by the sum of aggregates > 0.25 mm and classified as follows: <30%—unsatisfactory, 30–40—satisfactory; 40–75—good; and >75 as excessively high [
22]. The Kachinsky dispersion factor (DF) characterizes the degree of the destruction of aggregates in water and was calculated as the ratio of particles (<0.001) of “microaggregate” silt to “granulometric” silt (Equation (2)) [
22]. Soil penetration resistance was measured by repeating ten times from the soil surface to a depth of 45 cm in 2.5 cm intervals by using a soil compaction meter FieldScout SC 900 (Spectrum technologies, Aurora, IL, USA) equipped with a metal rod with a cone (size 1.3 cm). The assessment of the water-physical properties was performed for the non-eroded and pristine soil samples.
where numbers are the sizes of the fractions (mm). In the numerator, the sum of fractions is from 0.25 to 10 mm; in the denominator, the sum of fractions is >10 and <0.25 mm.
where
a is the silt content in the microaggregate analysis (%) and
b is the silt content in the particle size analysis (%).
Soil chemical analyses were carried out using the standard methods reported in [
24,
25]: the carbon (C) content, using the Tyurin method with termination according to Orlov and Grindel (Walkley-Black’s analogue); available phosphorus (P
2O
5) and exchangeable potassium (K
2O), according to Chirikov; exchangeable cations (Ca
2+ and Mg
2+) by the trilonometric method; and soil reaction by potentiometry (at 1 mol/L KCl suspension (1:2.5 soil/solution)). The content of available forms of ammonium (N–NH
4) and nitrate nitrogen (N–NO
3) were determined using a KCl solution. The gradation of SOM on the categories was carried out according to the scale [
26], where content > 10% is characterized as “very high”, 6–10%—“high”, 4–6%—“average”, 2–4%—“low”, and <4%—“very low”.
13C-NMR spectroscopy was used to study HAs for five samples: arable (two eroded and two non-eroded) and one pristine sample from a forest windbreak. HAs from the aforementioned samples were extracted according to a published IHSS protocol [
27] with modification by Vasilevich [
28]. The initial soil samples, sieved through a 1 mm sieve, were decalcified in a solution of 0.05N H
2SO
4 for one day. Then, after filtration, the decalcified solution was poured out, and the soil-solution of 0.1 N NaOH was poured into the sample with soil in a ratio of 1:10 and left for one day. Afterward, the supernatant was decanted and a coagulant (saturated solution of Na
2SO
4) was added. The next day, the solution was filtered again. To precipitate HA from the solution, a 1 N H
2SO
4 solution was used in a ratio of 50 mL of acid/100 mL of supernatant and left for a day. The HA gel was collected in plastic (dialysis) bags and placed in distillate water for 7 days. The water in the distillate tanks was changed every day. After dialysis, the gel was placed on Petri dishes and dried at room temperature.
Solid-state CP/MAS 13C-NMR spectra of the HAs were measured with a Bruker Avance 500-NMR spectrometer (Billerica, MA, USA) in a 3.2-mm ZrO2 rotor. The magic angle spinning speed was 20 kHz in all cases, and the nutation frequency for cross-polarization was u1/2p 1/4 62.5 kHz. Repetition delay was 3 s. The number of scans was 6500–32,000. The contact time used was 0.1–0.75 ms.
Table 1 shows the molecular fragments that we identified by CP/MAS
13C-NMR spectroscopy: carboxyl (–COOR), carbonyl (–C=O), CH
3–, CH
2–, CH–aliphatic, –C–OR alcohols, esters and carbohydrates, phenolic (Ar–OH), quinone (Ar=O), aromatic (Ar–).
The reliability of changes in the soil properties was assessed using the Student’s test. The significant differences between the soil properties were assessed by the least significant difference (LSD) of analysis of variance (ANOVA). Differences at the
p < 0.05 level were reported as statistically significant. Statistical analysis including mean and standard deviation values was performed using R 4.0.4 [
29] and RStudio (version 1.3.1093, Boston, MA, USA) [
30].
4. Discussion
The studied soils have undergone changes as a result of long-term conventional tillage, fertilization, and the application of various ameliorative and agronomic practices. Deterioration in the water-physical properties due to conventional tillage all over the world has been shown in numerous articles [
31,
32,
33,
34,
35] including for chernozems [
36,
37,
38]. For example, Trofimova [
39] previously showed that the number of agronomically valuable soil aggregates in the 0–30 cm soil layer of chernozems (ordinary and leached) under fallow and forest windbreaks exceeded the number under different systems of treatment: combined, shallow mulching, and zero tillage. Similarly, in the study of Schein [
40] on arable chernozems of the European part of Russia, the content of agronomically valuable soil aggregates under fallow land was higher, while the content of physical clay and silt fractions in the arable-fallow system was similar, which is comparable with our results.
Significant difference in the penetration resistance between the arable and virgin soil samples was due to prolonged exposure to heavy agricultural equipment and traditional tillage with a plough with a layer turnover. This anthropogenic impact led to the formation of a plough sole about 10 cm thick. The formation of a plough sole contributes to the deterioration of the physical properties of soils, disrupting the movement of water, nutrients and the development of erosion processes [
41,
42,
43]. Medvedev [
37] showed that medium- and heavy-loam soils are more prone to the formation of a plough sole. Since the surface soil layer is where most of the roots of crops occur (sugar beet and wheat), further deterioration by soil compaction will lead to growth limitation. For example, according to Kees [
44], at values already around 1500 kpa, there is a decrease in the root growth of most crops, while at 2500 kpa, the root growth of many plants stops.
In our study, the SOM content varied in the following direction: non-eroded cropland > forest windbreak > eroded cropland. All soils were characterized by “high” levels of SOM, despite erodibility and different types of land use. SOM content was lower under forest compared to arable non-eroded, which can be explained in different ways. The surface layer of soils under the forest is overlain by erosion material with a thickness of 2 to 10–15 cm transferred from the arable land due to eolian processes. The formed layer limits the flow of plant remains in the buried humus-accumulative horizon. Such conditions lead to processes of diagenesis of the labile part of the SOM and a slight decrease in its total content [
45,
46]. On the other hand, Baeva [
47] reported that when the natural-climatic zones changed from north to south, the difference between the organic C content in arable and pristine soils decreased, which may be related to the lower productivity of meadow vegetation in the steppe zones. Additionally, taking into account the high SOM content in arable non-eroded plots, we can conclude that chernozem soils are more resistant to anthropogenic influences [
48]. For example, no difference was previously found between the SOM content in fallow and arable plots of chernozems [
40]. Additionally, there is a possible influence on the maintenance of soil fertility by changing the composition of crops in the rotation [
49].
Reduced values of the SOM content in arable plots compared with pristine counterparts are characteristic of most soils as a result of conventional ploughing. A number of studies have reported a deterioration in fertility and a decrease in SOC content on arable chernozems in European Russia [
5,
6,
9]. A similar trend was observed on chernozems in the study region (Pre-Ural forest-steppe zone). Previously, it was found that the SOM content on virgin Haplic chernozems and Luvic chernozems exceeded their arable counterparts using turnover ploughing [
50].
Arable lands of the Pre-Ural steppe zone are subject to the influence of erosion processes, which limits their fertility [
21]. Involvement in arable farming with conventional tillage methods has contributed to the accelerated development of degradation processes, especially the intensification of processes of water and wind erosion. The average SOM values of eroded arable soil samples were lower than non-eroded samples by 1%, which is associated with the deflation and washout of silty and fine-grained soil fractions. The eroded soils were also characterized by a lower content of adsorbed cations.
According to
13C-NMR spectroscopy, the distribution of C in structural fragments in HAs (% of C total) corresponds to the previously studied chernozems [
51,
52,
53] including arable lands [
54] (
Table 6). Chemical shifts 110–160 (aromatic C) were the most pronounced of all the samples studied, caused mainly by the influence of lignin and tannins. We also did not exclude the degradation of the SOM and the selection of the most stable functional groups, in this case, aromatic fragments. However, this value was markedly lower (31) in the soil samples under the forest windbreak, which may be related to the processes of the humification of organic residues and the formation of aliphatic fragments, which prevailed over aromatic ones. However, it was previously noted that the content of aromatic fragments was slightly higher in the soils of the meadow ecosystems and significantly lower in the upper horizons of the soils of forest ecosystems [
55]. This is the result of the increased content of C alkyl forms in the HA molecules of soils formed under forest.
In our study, the differences between the arable and non-arable samples (under the forest windbreak) were revealed. It was found that the involvement of virgin chernozems in arable farming resulted in a decrease in aliphatic fragments. The AR/AL ratio was lower in the arable samples, which may be explained by microbiological destruction of the hydrolyzable aliphatic part of humic substances due to the conventional tillage system (ploughing) and, accordingly, the lack of conditions for the accumulation and decomposition of plant residues. Similar results were reported in other works on the study of arable chernozems. A number of studies have revealed a decrease in the amount of aliphatic chains and an increase in the content of aromatic structures and carboxylic groups [
15,
53]. Identical results have been reported in other climatic zones. For example, Lodygin and Abakumov [
11] reported that agricultural development of Eutric Albic Retisols (Loamic) led to a transformation in the HA molecular structure, which was demonstrated by a relatively higher fraction of aromatic molecule fragments and a decrease in the number of carbonyl groups. Thus, we can assume that long-term conventional tillage leads to an increase in the content of aromatic fragments and, accordingly, a decrease in aliphatic compounds.
The HAs of the eroded and non-eroded soils were similar, although a more marked difference was previously found between the eroded and non-eroded arable chernozems. It was shown that the soil samples not subjected to erosion had more aromatic fragments compared to the eroded soil samples [
54]. In general, the HA structure of the studied soils corresponds to the soils of the forest-steppe zone, indicating the similarity of the biothermodynamic conditions of humus formation in these soils [
56].
Switching to conservation tillage systems is seen as a rational way to improve soil fertility. Previous studies have concluded that reduced tillage enhances SOM stabilization during the transition from cropland to grassland and vice versa [
12,
48]. For example, Shrestha [
13] reported a predominance of O-alkyl C in soils under no-till compared to conventional tillage, suggesting a more advanced stage of SOM decomposition. Similarly, no-till promoted the formation of stable HAs with a higher proportion of aliphatic and hydrophobic compounds, while the hydrophobicity of aliphatic fragments in HAs improved the stability of C [
57]. Previously, it has been reported that the use of long-term conservation tillage and grain–fallow–grass crop rotation practices (zero till) improved the physical and chemical properties of soils and contributed to reducing the activity of erosion processes in the study region (Republic of Bashkortostan) [
50,
58]. Thus, the most expedient is considered to be the transfer of arable land to fallow land, or a change of agricultural practices to conservation practices. We imply that the transition to no-till tillage will reduce erosion, and improve the physical and chemical properties of agrochernozems and C sequestration.