Soil Health in an Integrated Production System in a Brazilian Semiarid Region

Abstract

Soil health is directly related to sustainable development goals (SDGs) and can be affected by inadequate management practices. In this work, soil edaphic respiration and changes in microbial biomass promoted by cover crops in an integrated crop–livestock system (ICLS) were evaluated using soil health indicators with the respirometry method. The design was completely randomized in a 3 × 6 factorial arrangement, and multivariate principal components analysis (PCA) was performed according to MANOVA. Edaphic respiration was determined based on the respirometry technique. The results showed that edaphic soil respiration was significant in the nine evaluation periods, demonstrating the importance of grass cover in edaphic respiration arising from the biological activity of microorganisms, which is directly related to the amount of soil organic carbon. The cover crops increased soil organic matter and consequently microbial respiratory activity.

1. Introduction

Pasture degradation and poor land use through traditional agriculture can compromise environmental, economic, and social sustainability, contributing to modifying the carbon and nitrogen cycles [1]. According to the sustainable development goals (SDGs), the use of technologies such as minimal soil tillage and crop rotation plus integrated crop–livestock systems (ICLSs), which favor degraded pasture recovery and improve straw production, besides physical, chemical, and biological soil properties, re-emerges as a consolidated strategy for sustainable intensification in harmony with environmental concerns [2,3,4,5].

Soil is an essential component in the life maintenance process, since it promotes the dynamics and storage of water and maintains food chains, the environmental regulatory functions, nutrient cycling, and the diversity of macro- and microorganisms, representing the main life regulation element [6]. Land-use intensification associated with unsustainable management practices can affect soil quality, reducing its density, fertility, and biological activity [7,8].

As soil is the basis for sustainable production, soil health is directly related to an agroecosystem’s functions [9,10,11]. Soil health is the persistent ability of soil to function as a vital ecosystem that sustains plants, animals, and humans, including soil biota/biodiversity, and related soil functions and soil-based ecosystem services [12]. Thus, cover crops have been used as an ecological and economical alternative to provide benefits to soil, including increases in physicochemical and biological parameters [13], thus contributing to form organic matter (OM) to protect soil [14], as well as attracting edaphic organisms by offering shelter and food [15,16,17].

Physical and chemical soil attributes are involved in the critical functioning of soil and can serve as indicators for healthy-soil assessment. Among the indicators that characterize the physical properties of soil, soil texture, bulk density, and porosity assess aeration, retention of nutrients and water, and soil erosion. At the same time, acidity, exchangeable cations (i.e., Ca²⁺, Mg²⁺, and K⁺), total organic carbon, and available phosphorus are indicators of high sensitivity to environmental transformations [18]. However, because management practices have only limited short-term effects on some of these soil attributes, it is also necessary to explore biological indicators such as biomass and basal respiration [19]. Thus, the evaluation of soil microbial activity has been proposed as an indicator of soil quality due the soil microorganisms decompose plant debris into transformed carbon products in the soil. Therefore, it is a more sensitive indicator of soil quality than any other soil physical and chemical parameters and thus more apt to monitor environmental changes resulting from agricultural use [7,20].

Soil microbial activity is largely influenced by temperature, pH, luminosity, salinity, energy sources, organic substrates, and nutrients, besides the presence or lack of toxic elements. Since biological activity is substantially higher in the surface soil layer, vegetation removal affects the surface microbial community, generally decreasing the diversity, evenness, and richness [21].

Soil respiration, usually defined as the CO₂ produced by biological activity in the soil, is a strong tool to evaluate decomposition intensity [22], as it represents the biological activity of organic waste [23] and can document changes in soil carbon dynamics in deforested areas for croplands [15]. In this context, an appreciable body of research has used soil properties to understand land-use effects on the edaphic environment [12].

Understanding degradation processes and assessing soil quality require a multivariate approach, along with physical, chemical, and biological parameters under different usage and management conditions [24]. The respirometry method is a cost-effective and easy-to-perform technique that estimates soil microbial activity [25]. Thus, the aim of this study was to evaluate soil edaphic respiration in an ICLS and the alterations promoted by cover crops in their microbial biomass.

2. Materials and Methods

2.1. Experimental Site

The study was conducted in two stages from March to July 2018 in an experimental site of Embrapa Algodão at the Experimental Station of State Agricultural Research Corporation of Paraíba (EMPAER). The site is located in the municipality of Lagoa Seca (07°10′15″ S, 35°51′13″ W. Gr., altitude of 634 m), Mesoregion of Paraíba, Microregion of Campina Grande. The region’s climate is tropical with dry summer (As), with annual average temperature of 22.5 °C and annual rainfall of 835 mm [26]. The soils are classified as Neossolos Litólicos Distróficos típicos [27] or Eutric Leptosol Arenic [28], formed from the alteration of orthogneisses and granodiorite migmatites.

Soil samples were collected in an area with five years of consolidation under a low-carbon agricultural production system. This area comprised the tested integrated crop–livestock systems (ICLSs).

Various combinations of grasses (corn and sorghum) and legumes (pigeon peas, and crotalaria or rattlepods) were associated with different forage grass species as soil cover, totaling 25 ICLS treatments (

Table 1. Treatments in the experimental area of Embrapa Algodão/EMPAER.

Manual Planting Together with Corn
1	Brachiaria brizantha cv Piatã
2	Brachiaria brizantha cv Marandu
3	Urochloa mosambicensis—Urochloa grass
4	Cenchrus ciliaris (L.)—buffel grass
5	Brachiaria decumbens
6	Panicum maximum cv Massai
Planting between Rows along with Corn
7	Brachiaria decumbens
8	Brachiaria Brizantha cv Paiaguás
9	Brachiaria Brizantha cv Piatã
10	Corn + Brachiaria + Stylosanthes
11	Corn + Piatã + Stylosanthes
12	Mombaça by hand
Planting between Rows 14 Days after Corn
13	Brachiaria Brizantha cv Piatã
14	Brachiaria Brizantha cv Paiaguás
15	Brachiaria brizantha cv Marandu
Planting 14 Days after Sorghum Grain
16	Panicum maximum cv Massai
17	Urochloa mosambicensis
18	Brachiaria Brizantha cv Piatã
19	Brachiaria Brizantha cv Paiaguás
20	Panicum maximum cv Mombaça
Planting 14 Days after Corn by Hand
21	Panicum maximum cv Massai
22	Urochloa mosambicensis—Urochloa grass
23	Brachiaria Brizantha cv Piatã
24	Brachiaria Brizantha cv Paiaguás
25	Panicum maximum cv Mombaça

).

A completely randomized design was used in a 3 × 6 factorial arrangement with three soil collection depths (SD; 0–10, 10–20, 20–30 cm), five vegetation cover types, and the control (without cover), besides three replications, totaling 54 experimental units. The species composing the treatments were those planted 14 days after corn sowing, namely, Brachiaria brizantha (Piatã and Marandu cultivars, Paiaguás); Urochloa mosambicensis (Urochloa); Cenchrus ciliaris (buffel grass); and Panicum maximum (Massai and Mombaça cultivars) (Table 1). The control treatment was an adjacent area with no vegetation cover.

2.2. Soil Sampling and Collection

The first trial stage consisted of collecting soil material (samples) from the five vegetation cover types: Brachiaria brizantha (Piatã and Marandu cultivars, Paiaguás); Urochloa mosambicensis (Urochloa); Cenchrus ciliaris (buffel grass); and Panicum maximum (Massai and Mombaça cultivars). Vegetation and residues were removed from the soil surface and classified as Dystric Leptosol Arenic [28]. These soils present continuous rocks starting ≤ 50 cm from the soil surface and are rich in coarse fragments. The samples were collected at three depths (0–10, 10–20, 20–30 cm) using a Dutch auger, and five simple samples were taken from each depth to form a composite, totaling 90 samples packed in plastic bags.

Composite samples for physical and chemical soil analyses were sent to the Laboratory of Soil Chemistry and Fertility of the Center for Agricultural Sciences (CCA), Federal University of Paraíba (UFPB), Campus II, in the municipality of Areia, State of Paraíba (

Table 2. Chemical characteristics of the soil used in the experiment.

Attributes
pH	P	K	Na+	H + Al+3	Al+3	Ca+2	Mg+2	OM
H2O	mg dm−3	-------------------------cmolc dm−3---------------------						g dm−3
6.2	45.5	65.1	0.0	3.22	0.05	0.40	0.40	7.05

).

After soil samples were collected, they were air-dried in a greenhouse for 72 h. Then, they were ground to pass a 2 mm sieve to obtain fine-earth fraction samples (FEFSs). The samples were taken to the Laboratory of Soil Chemistry and Fertility, UEPB, Lagoa Seca-PB, and dried in a forced-air oven at 65 °C for 72 h to obtain dry soil at constant weight.

Physical and chemical soil analyses were performed using standard methods from Brazil [29]. Particle density was determined using the volumetric flask method. The total porosity was obtained indirectly by measuring the bulk density. The pH was determined in water (1:2.5—FEFS:H₂O). Exchangeable contents of Ca²⁺, Mg²⁺, and Al³ were extracted with 1 mol L⁻¹ KCl, while available P, K⁺, and Na⁺ were extracted with Mehlich 1 solution (0.05 mol L⁻¹ HCl + 0.0125 mol L⁻¹ H₂SO₄). The potential acidity (H + Al) was extracted with 0.5 mol L⁻¹ Ca(C₂H₃O₂)₂ (pH 7.0). Ca²⁺ and Mg²⁺ contents were determined using complexometry; Al³⁺, using titration; K⁺ and Na⁺, using flame photometry; and P, using colorimetry. Total organic carbon (TOC) was determined according to the Walkley–Black chromic acid wet oxidation method. Cation exchange capacity (CEC) and base saturation (BS) were also calculated [29].

Regarding the microbial respiration analysis, the soil water-retention capacity was determined using the funnel method with the previously dried soil distributed in a 250 mL Erlenmeyer flask. Soil moisture was maintained at 60% of field capacity. Soil samples of the respective treatments were then placed in transparent plastic pots of 0.5 L to determine edaphic respiration, in which 0.2 kg of soil was placed, and the alkali solution was allocated in a pot with a volume of 40 mL in the amount of 25 mL of NaOH (0.2 N). The technique was determined by measuring the difference in the acid volume required to neutralize the sodium hydroxide contained in the glasses [30].

The containers were opened at four-day intervals (nine readings) and titrated with HCl (2N) with phenolphthalein acid/base indicator; they were then evaluated at the Soil Chemistry and Fertility Laboratory for titration with HCl acid (0.2 N) in a 25 mL automatic pipettor using three drops of phenolphthalein as an indicator. After reading, another 25 mL of HCl solution (2N) was added; then, the containers were closed. The difference in the acid volume needed to neutralize sodium hydroxide in the treatment was proportional to the amount of carbon dioxide produced by soil microorganisms.

Edaphic respiration was calculated using the following formula [31] (Equation (1)):

CO₂ = (V1 − V0) × 44 ÷ 0.2

(1)

where CO₂ is the amount of mineralized carbon (mg of CO₂ kg⁻¹ of soil), V1 is the volume of HCl needed to neutralize NaOH in the treatment (mL), V0 is the volume of HCl needed to neutralize the control (mL), 44 is the CO₂ molar weight equivalent, and 0.2 is the mass of soil (kg).

2.3. Statistical Analysis

After confirming the data’s normal distribution, the dataset was subjected to the following analyses: (i) multivariate analysis of principal components (PCA) to verify the distinction degrees of management types and the possible associations between them and the variables and (ii) hierarchical cluster analysis. R software (version 4.1.0) was used for such analyses [32].

3. Results

The principal component analysis grouped the variables into three principal components (PC₁, PC₂, and PC₃) that had eigenvalues greater than 1 (λ > 1.0). The first three PCs explained 91.22% of the total accumulated variance, with PC₁ accounting for 68.92%, PC₂ explaining 16.22%, and PC₃ contributing 6.07%. Conversely, PC₄ characterized a univariate process related only to soil phosphorus content (

Table 3. Summary of principal component analysis, and multivariate variance (MANOVA), and univariate (ANOVA) analyses.

Indicators	Principal Components
	PC1	PC2	PC3	PC4 **
	Pearson Correlation Coefficients (r)
R1—Microbial respiration at 4 days	0.98 *	−0.09	0.03	0.07
R2—Microbial respiration at 8 days	0.97 *	−0.23	−0.04	0.07
R3—Microbial respiration at 12 days	0.96 *	−0.17	−0.05	0.02
R4—Microbial respiration at 16 days	0.96 *	−0.18	−0.10	0.03
R5—Microbial respiration at 20 days	0.97 *	−0.17	0.05	0.03
R6—Microbial respiration at 24 days	0.90 *	−0.26	0.33	0.04
R7—Microbial respiration at 28 days	0.94 *	−0.17	0.28	0.03
R8—Microbial respiration at 32 days	0.97 *	−0.20	0.13	−0.01
R9—Microbial respiration at 36 days	0.98 *	−0.18	−0.02	−0.04
pH—hydrogen potential	0.67 *	−0.08	−0.66 *	0.18
P—phosphorus content in soil	0.00	0.64 *	0.07	0.75 *
Al—aluminum content	−0.93 *	0.20	0.16	0.06
OM—organic matter content	0.58 *	0.80 *	−0.11	−0.02
CEC—cation exchange capacity	0.64	0.64 *	−0.02	−0.34
V%—base saturation	0.91 *	0.28	−0.09	0.09
PD—particle density	−0.55 *	−0.52 *	−0.54 *	0.01
TP—total porosity	0.52	0.77 *	−0.19	−0.29
λ—eigenvalues	11.72	2.76	1.03	0.83
σ2 (%) total explained variance	68.92	16.22	6.07	4.86
σ2 (%) total accumulated variance	68.92	85.14	91.22	96.07
Variation sources	Wilk Test (p-Value)			F-Test (p-Value)
Var—soil cover varieties	<0.01	<0.01	0.01	<0.01
SD—sampling depth	<0.01	<0.01	0.01	<0.01
Var × SD—interaction between factors	<0.01	<0.01	0.01	<0.01

*: correlation coefficients higher than 0.5 considered in the principal components; **: single variable in the principal component subjected to analysis of variance using the F-test.

). The first three PCs and phosphorus content were significantly influenced by soil cover, depth, and their interaction.

According to the MANOVA, basal soil respiration was influenced by nine evaluation periods, demonstrating the importance of cover (grass) in edaphic respiration arising from microbiological activity directly related to soil organic carbon content (Table 3). Generally, microbial biomass carbon (MBC) represents 1 to 4% of the TOC. Also, qMIC values below 1% can be attributed to some limiting factor of microbial biomass activity [7,20,33].

With the two-dimensional graphic of PC scores (Figure 1A) and Pearson correlation coefficients between the PCs and the original variables (Figure 1B), we could verify a general split in the factors (grass varieties and depth) into two components. PC1 represented 68.92% (grass) of the variance and was correlated with the grass factor, while PC2 described 16.22% of the variance and reflected the sampling depth factor. Grass varieties and soil sampling depths were distributed across the four quadrants of the PCA figure (Figure 1A). Coverage systems with higher grass diversification and soil sampling depth were grouped in the second and fourth quadrants (p > 0.1). This established five associations with the indices and groups of cover varieties (grass), and sampling depth, while the others were separated by the first and third quadrants (p > 0.1) (Figure 1A). Attributes such as soil microbial biomass have been successfully used to assess soil quality and measure disturbances in a given environment, which helps determine the sustainability of agricultural practices [5]. In this sense, soil basal respiration measures the microbiological activity of soil where microorganisms degrade organic compounds to CO₂, thus an important and sensitive indicator of soil quality activity [7,20].

Among the associations formed between cover crops and sampling depths in the second and fourth quadrants, PC₁ showed that BRS Paiaguás and BRS Piatã grasses stimulated higher microbial respiratory activity (R1 to R9) at 0–10 cm (Figure 1A,B).

The original soil edaphic respiration confirm the PCA results. However, the mean edaphic respiration values were reduced as a function of sampling depth, regardless of soil cover. This higher soil microbial respiratory activity is directly related to soil management through the integrated crop–livestock system (ICLS) with BRS Paiaguás, BRS Piatã, and Massai grasses as soil cover, providing an increment in straw on the soil surface, consequently increasing SOM content after decomposition processes. Previous research assessing the influence of different systems on microbial activity has found no significant differences concerning basal soil respiration among integrated crop–livestock systems, native vegetation, or native vegetation in recovery [34].

The averages of soil physical and chemical properties confirm the PCA results. PC₃ showed that soils covered with BRS Paiaguás and Urochloa grasses had a higher hydrogen-ion concentration (pH) than those with BRS Piatã and Massai. The adoption of minimal soil disturbance increases soil organic matter and generally results in higher hydrogen-ion concentration due to the production and release of organic acids throughout the mineralization process.

The oxidation process releases electrons into the soil solution under organic matter accumulation conditions in the final mineralization stage, which increases pH, even at depth [35]. No-tillage systems with several years of stabilization also present lower toxic aluminum, possibly due to the complexing role of Al⁺³ in soil organic matter. Conventional tillage presents higher acidity than the no-tillage system [36].

PC2 showed that Mombaça grass cover provided higher phosphorus (P) content in the soil. This available P content decreased with the increase in soil depth. The higher P content in the superficial soil layer is due to root residue decomposition from cover crops and grown plants such as sorghum and others added to the system, which uses P fertilizer for its development.

Higher P levels in the no-tillage system are attributed to minimal soil disturbance [37]. Similar to P content, the use of BRS Paiaguás and BRS Piatã grasses as mulch significantly increased base saturation and soil cation exchange capacity (CEC) (Figure 1A,B). Increases in base saturation and CEC in an ICLS with such grasses are due to their potential as straw producers to enhance soil organic matter. Research carried out under different climatic conditions has shown that no-tillage systems improve base saturation compared with the conventional system conducted with monocultures over five years due to higher organic matter, magnesium, calcium, potassium, and cation exchange capacity [38]. Furthermore, the gneiss structure (parent material) controlling water drainage, retention, and redistribution throughout the soil layers contributes to the formation of kaolinite and smectite in the clay fraction of these soils, contributing to high CEC values and plant nutrient retention [39].

Low contents of organic matter on surface soil with low-activity clay in no-tillage systems are sufficient to increase the effective and potential CEC, with significant effects up to 8 cm deep [40]. Thus, the SOM in these systems plays an important role in maintaining the CEC of soils with predominance of mineral fractions with lower reactivity. Also, the decrease in soil organic matter content in traditional crops reduces CEC [33].

Soil cover with BRS Paiaguás and BRS Piatã grasses reduced the aluminum (Al⁺³) content compared with the soil without vegetation cover (control). This reduction in Al⁺³ content is due to its complexation with negatively charged acidic functional groups on soil organic molecular surfaces. Organic matter has a high concentration of functional groups, among which are the carboxylic groups that complex Al through electrostatic attraction, which allows Al to be more easily leached, since metal–humus complexes are primarily formed through the interaction of metal and carboxylic functional groups [41].

The role of OM in reducing Al⁺³ through complexation has already been previously demonstrated with the application of cattle manure and poultry litter as an alternative for fertilization in Eutric Leptosol Arenic [42]. The amount of organic residue in soil directly affects soil biomass. The exudation of organic acids [15,43], such as lactic, acetic, citric, maleic, oxalic, tartaric, and succinic acids, originates from a variety of sources, which include plant roots, microorganisms, and organic decomposition. Organic acids can participate in complexation reactions of aluminum ions, reducing their toxicity to plants and buffering soil pH [44,45].

Regarding SOM content (PC2), the highest OM contribution occurred in the most superficial soil layer (0–10 cm) when Mombaça grass was grown as soil cover. However, SOM significantly decreased in deeper layers (Figure 1A,B).

Improvements in soil biological activity resulted from the adding Mombaça grass as straw. Panicum maximum and Urochloa spp. are satisfactory sources of straw for NTS due to significant dry matter yield [46]. According to the two-dimensional graphic (Figure 1C) and the eigenvectors (Figure 1D), a significant effect of the cover (grass) on particle density and total soil porosity (soil physical attribute) was observed.

The associations found (PC₃) between cover and soil depth for soil physical attributes were observed in the third quadrant (Figure 1D). Thus, BRS Paiaguás, Mombaça, and BRS Piatã grasses presented the lowest particle density compared with the control, with particle density increasing at depths. Urochloa showed the lowest efficiency in reducing particle density and total porosity among the evaluated grasses.

OM forms macroaggregates in soil, improving its aggregation and total porosity and consequently enhancing soil water storage and aeration. Previous research has showed increases in aggregate stability and water infiltration rate, as well as decreases in bulk density and compaction in ICLSs [4]. The dendrogram obtained with hierarchical cluster analysis is shown in Figure 2. Soil cover forms the G1 cluster, while the G4 cluster grouped three cover types, V2-BRS Paiaguás, V3-Urochloa, and V4-Mombaça, at different soil depths (Figure 2A).

The hierarchical clustering of PC2 showed three distinct groups (G1, G2, and G3), where G3 exhibited the smallest number of grasses (V2 and V4) and G1 presented the greatest number of grass genotypes combined with soil depths (Figure 2B).

4. Discussion

Higher CO₂ release often occurs due to increased biological activity directly related to soil labile carbon [47]. The MANOVA demonstrated that grass cover and depth significantly affected the first three PCs, including physical and chemical soil attributes (pH, P, Al³⁺, and OM contents, besides CEC, base saturation, particle density, and total porosity). Thus, our results confirm an interaction between the soil cover and sampling depth factors on edaphic respiration, and soil physical and chemical properties. This effect can be attributed to the permanent grass cover in no-tillage systems, as their root system can reach much greater soil depths, improving soil physical, chemical, and biological properties (Table 3).

No-tillage systems promote the formation of a microenvironment with higher organic soil carbon content that favors soil moisture and fertility maintenance, facilitating P diffusion into soil solution and its absorption by plants [48]. In addition, the use of ground-cover plants can accelerate the release of water-soluble organic acids that form complexes with exchangeable aluminum, which reduces its phytotoxicity and increases the adsorption calcium, magnesium, and potassium in soils, thereby preventing their leaching [2,3,5,32,49,50].

Evaporation losses in soil with straw are lower than those in soil without cover, promoting a more suitable environment for crop establishment [51]. Previous research evaluating OM compartments under different covers has found lower total organic carbon (TOC) levels at deeper soil layers and higher TOC levels in native forest soil than in cropped soil [52]. These observations confirm higher reserves and contributions of OM in forest soils.

Cover crops, usually grasses like Mombaça, present roots with a good ability to penetrate deeper layers, with significant effects on nutrient absorption and biomass production. The stock of high contents of P in biomass enhances P cycling, contributing to its transfer to upper surface layers. Cover crops may increase P at the soil surface, given its low availability due to slow diffusion and high fixation in soil. Thus, our results confirm higher P cycling in no-tillage systems than in the conventional system. This is similar to previous studies that evaluated long-term P levels in no-tillage and conventional management systems [37].

No-tillage systems improve the chemical attributes of soil due to minimum soil disturbance. Different root systems contribute to increasing OM through root colonization in deep soil layers. As a result, SOC increases in depth, which benefits other functions of soils. Such improvements include enhanced soil aggregation, microbiological activity, soil water storage, and infiltration rate, and reduced erosion, with positive effects on aggregate stability, porosity, and soil density [53]. Due to its low density, OM contributed to the increase in soil total porosity in ICLSs compared with the control (with no vegetation cover). Previous research has reported that integrated systems increase aggregate stability and water infiltration rate and decrease bulk density and compaction [4].

Leaves and branches on the soil in forests provide better physical soil conditions. The decomposition of these materials increases SOM content, diminishing bulk density and improving soil aggregation [54,55,56]. Also, divergent root systems due to multiple species enhance the OM levels, since soil colonization by roots is a manner to increase OM in deeper layers, improving the structure and creating biopores [15,57].

5. Conclusions

BRS Paiaguás and BRS Piatã grass varieties promoted higher soil microbiological activity at 0–10 cm, followed by Massai and Urochloa grasses. This microbial activity constitutes a crucial component of soil health, acting as a biological indicator for the assessment of sustainability and productivity in agroecosystems. Thus, the use of these grasses improves soil physical, chemical, and biological attributes and promotes higher infiltration and available water storage capacity, higher soil aggregate stability, and significant input of organic carbon and nutrient cycling. However, it is necessary to continue researching to verify improvements in soil health indicators.