Evaluation of the Effects of Different Cultivars of Falcataria falcata on Soil Quality

Abstract

The soil microbial community influences and maintains soil quality and health. Leguminous plants are widely used in forestry due to their nitrogen-fixing ability, significantly improving soil quality. However, there are few studies on the effects of woody legumes on soil microbial communities and soil quality. Here, the composition and structure of bulk soil microbial communities associated with six cultivars of Falcataria falcata (L.) Greuter & R. Rankin were analyzed using full-length 16S rRNA sequencing. Additionally, the minimum dataset approach was employed to integrate indexes of soil microbial communities and physicochemical properties, allowing the calculation of a soil quality index to evaluate the cultivars’ soil quality. Although the growth characteristics of the six cultivars were identical, there were significant differences in physiological functions. Notably, cultivar 6 demonstrated a significant advantage over the other cultivars in its overall physiological characteristics. Compared to the control sample, all cultivars clearly improved soil quality, with cultivars 8 and 16 significantly outperforming the others. The findings indicate that the different cultivars improve soil fertility by recruiting microorganisms with specific functions. The stability of soil microbial communities is a crucial biological and ecological factor that influences and sustains soil quality and health and is a key index for the evaluation of these properties.

1. Introduction

In recent decades, concentrated economic development and accelerated industrialization in various countries have reached a critical turning point, posing significant threats to both soil and human health. These threats include soil degradation, low efficiency of resource utilization, and environmental deterioration. Additionally, improvements in living standards and rising urbanization rates have further contributed to the decline in soil quality [1,2]. Consequently, land degradation has emerged as a major global challenge, particularly in arid and semi-arid regions, where it results in soil erosion, decreased soil fertility, and reduced biodiversity [3]. In China, land degradation significantly impacts the provision of ecosystem services, particularly in vulnerable regions such as the Loess Plateau, where prolonged cultivation has intensified soil degradation and erosion [4]. To combat these challenges, various policies and land protection programs focused on ecological restoration have been implemented, including the United Nations Decade on Ecosystem Restoration [5] and China’s Grain-for-Green Program [6], aimed at increasing ecosystem diversity and enhancing soil carbon sequestration capacity while improving soil quality. Soil quality refers to the ability of soil to sustain plant productivity, water, and air quality and to protect the health of humans, plants, and animals, as well as habitats [7], and is important to maintain healthy soil [8]. Therefore, determining how to evaluate, improve, and maintain the quality and health of soil has become an important and urgent task for current workers in agroforestry and ecology.

Soil is a highly complex ecosystem, harboring the most diverse microbial communities on Earth, including bacteria, archaea, fungi, viruses, protozoa, and various microfauna, collectively referred to as the soil microbiome [9]. The soil microbiome plays a crucial role in the biogeochemical cycles of organic matter and various elements, such as nitrogen and phosphorus, as well as in the degradation of soil pollutants [10,11,12]. Consequently, it is closely related to soil health and crop production. An increasing number of researchers believe that, with regard to soil health, the main considerations should be the microbial components of the soil and the functions of soil ecosystems, particularly those related to energy flow, material cycling, and information exchange within the system [8,10,13]. However, due to the complexity of soil ecosystems and the limitations of current research technologies, the relationship between microbiological characteristics and soil ecosystem function remains unclear, which hinders the development of soil health indexes based on the soil microbiome [8].

With breakthroughs in high-throughput sequencing technologies and advances in bioinformatics, research on soil microbiomics has progressed rapidly. For instance, researchers can identify microbial interactions in specific habitats through co-occurrence network analysis [14]. The stability of soil ecosystems can be predicted by analyzing the interactions among microbial community members via co-occurrence network methods [15,16,17]. This in-depth understanding of microbial interactions is essential for maintaining and evaluating soil quality. Therefore, it is now technically feasible to incorporate soil microbiome indexes, alongside soil physicochemical indexes, as significant criteria for soil quality assessment, so that a comprehensive soil quality assessment system can be constructed. This integrated soil quality assessment system, which combines biological and physicochemical markers, will more accurately reflect the actual quality and health of the soil. Such assessments will not only help to improve agroforestry management but also support policy making to ensure soil versatility and the delivery of ecosystem services [18,19]. Furthermore, these assessments contribute to the enhancement of environmental quality while supporting the stable economic development of rural areas [20,21].

Although soil microbial communities initially form a “microbial seed bank” in the rhizosphere, differences in the composition of these communities in the presence of different plant species, and even for a single species as it develops, suggest strong host/plant-specific selection of soil microorganisms [22,23,24], which may be related to the nature of plant root exudates. Plants transport carbon (C) fixed during photosynthesis to the rhizosphere through root exudation, a process crucial for regulating the abundance and activity of soil microorganisms. Root exudates primarily consist of a variety of labile low molecular weight (LMW) carbon sources, which provide easily accessible energy for soil microorganisms and promote their recruitment [25,26,27]. The production and composition of these root exudates are strictly controlled in plants by genetic factors [28,29,30]. Therefore, the selection of suitable plants or cultivars (varieties) is essential to achieving ecological and economic goals in agriculture and forestry [31,32]. Different cultivars (varieties) of the same plant exhibit significant differences not only in growth characteristics and yield but also in their potential to improve soil quality and enhance ecosystem services [33,34]. Although the importance of cultivar (variety) selection is widely recognized [35], the understanding of differences in the mechanisms of interaction between various plant varieties and soil is still limited, which inhibits the development of cultivar breeding and soil improvement and management strategies.

Falcataria falcata (L.) Greuter & R. Rankin, a tall evergreen leguminous tree, is native to Malacca in Malaysia and the Maluku Islands in Indonesia; it is one of the fastest-growing tropical and subtropical nitrogen-fixing trees and is vigorously advocated and promoted by the International Union of Forest Research Organizations (IUFRO) [36]. Additionally, it is an economically valuable species from which timber and paper can be produced. This tree species was introduced to southern China around 1940 and has been widely cultivated in Guangxi and Guangdong provinces [37]. Up to now, there have been very few studies on the growth, development, and nutrient requirements of F. falcata [38]. Furthermore, there are no reports on how different cultivars of F. falcata affect soil microbial communities and soil quality. To address this issue, we selected for this study six cultivars of F. falcata that are promoted in Guangdong. Using traditional soil physicochemical testing methods alongside modern high-throughput sequencing techniques and bioinformatics methods, we monitored and analyzed the bulk soil physicochemical properties, microbial communities, and characteristics of plant growth and development during the early stages of cultivation. We also aimed to integrate soil microbial indexes into physicochemical indexes to assess soil quality, providing theoretical references for forestry production and ecological protection. This study seeks to clarify the following scientific questions: During the early stages of F. falcata cultivation, (1) are there differences in physiological growth characteristics, the physicochemical properties of the bulk soils, and microbial communities among different cultivars? (2) What are the key microbial factors that influence and evaluate the bulk soil quality of different cultivars? (3) Which cultivars are more suitable for ecological cultivation in forestry and for economic timber use?

2. Materials and Methods

2.1. Sample Collection Locations and Strategies

The six cultivars (3, 4, 5, 6, 8, and 16) of F. falcata were selected for this study from experimental demonstration forests, which have grown for over two years without human interference since their initial planting, covering an area of 1.333 hectares, with a plant and row spacing of 3 m × 3 m. They were located in the state-owned Meihua Forest Farm, Boluo County, Huizhou City, Guangdong Province, at an altitude of 84.5–108.7 m, 114°15′25″ E and 23°12′48″ N, with an average annual temperature of approximately 21.9 °C, average annual precipitation of 1940 mm, and annual actual evapotranspiration of 946.45 mm, all of which are characteristic of a South Asian subtropical monsoon climate. This type of soil is lateritic red soil [39]. To systematically assess the growth and development of each cultivar, as well as the soil conditions, nine plants of each cultivar were randomly selected for the measurement of growth and physiological parameters. Additionally, bulk soil samples were collected for analysis. This approach ensured the representativeness of the samples and balance in the experimental design. Given that F. falcata is a shallow-rooted tree species with a well-developed lateral root system, it is primarily concentrated in the 0–20 cm soil layer. Therefore, after removing surface litter and dead branches (Litter layer), a wooden spade (sterilized with 70% alcohol prior to use to prevent cross-contamination) was employed to extract the loose soil (Humus layer) around the root zone of each plant [40]. Soil samples were then collected 0–20 cm below the topsoil at each of the four cardinal directions (east, south, west, and north) around each plant. The soil samples were passed through a 2 mm sieve to remove roots and stones, and then the samples from all four directions were thoroughly mixed to obtain approximately 1 kg of a composite soil sample, which was subsequently sealed in sterilized, high-temperature resistant, moisture-proof sterilization bags (45 × 55 cm) for further analysis. Additionally, a control group (CK) was established in a barren area devoid of vegetation near the demonstration forest, where nine soil samples were randomly collected. The collection of these samples adhered to the same specifications as that for the cultivar group. After collection, the samples were quickly transported back to the laboratory, where each soil sample was divided into three portions: one portion was rapidly placed in a −80 °C ultra-low temperature freezer for soil microbial DNA extraction, the second portion was air-dried, and the final portion of fresh soil was stored in a 4 °C refrigerator to measure soil physicochemical properties and enzyme activity.

2.2. Measurement and Comprehensive Evaluation of Physiological and Growth Index

The physiological and growth index was determined after taking into account the diameter of the plant at ground level, the diameter of the plant 1.3 m above the base of the trunk (‘breast height’), plant height, and root nitrogen and phosphorus concentrations, as well as net photosynthesis and stomatal conductance. The diameter at ground level was measured using a caliper, while the diameter at breast height was measured with a diameter tape. Plant height was measured from the ground to the top of the plant using a measuring pole. Before measuring root nitrogen and phosphorus concentrations, the roots were rinsed with tap water to remove surface soil and impurities. They were then rinsed again with deionized water to ensure that all residual contaminants were eliminated. The cleaned root samples were placed on clean filter paper to drain excess water and were subsequently dried at 60 °C until a constant weight was achieved. After drying, the roots were chopped and digested. Root nitrogen and phosphorus concentrations were then determined using the Kjeldahl method and the molybdenum-antimony colorimetric method [41], respectively. The net photosynthetic rate and stomatal conductance were measured directly in the field using a Li-6400 portable photosynthesis system (Li-6400, Li-Cor, Lincoln, NE, USA). The physiological and growth index values were derived from these measurements and comprehensively evaluated using the fuzzy membership function method in R (version 4.3.0) [42,43]. The membership function is calculated as follows:

$${\widehat{\mathrm{X}}}_{\mathrm{ij}}={\displaystyle \frac{\left({\mathrm{X}}_{\mathrm{ij}}-{\mathrm{X}}_{\mathrm{jmin}}\right)}{\left({\mathrm{X}}_{\mathrm{jmax}}-{\mathrm{X}}_{\mathrm{jmin}}\right)}}$$

(1)

where ${\widehat{\mathrm{X}}}_{\mathrm{ij}}$ is the membership function value of the physiological and growth index j in cultivar i; X_ij is the actual value of the physiological and growth index j in cultivar i; X_jmin is the minimum value of the physiological and growth index j in the evaluated cultivars; X_jmax is the maximum value of the physiological and growth index j in the evaluated cultivars.

2.3. Determination of Soil Physicochemical Properties and Enzyme Activity

The physicochemical properties of soil include pH, soil water content (SWC), nitrate nitrogen (NO₃⁻-N), ammonium nitrogen (NH₄⁺-N), available potassium (AK), available phosphorus (AP), total nitrogen (TN), and soil organic carbon (SOC). Soil pH was measured using a pH meter (DELTA 320) with a soil-to-water ratio of 1:2.5 [44]. SWC was determined by weighing 20 g of soil samples after drying them in an oven at 105 °C for 24 h [45]. NO₃⁻-N was extracted using 1 M potassium chloride and subsequently measured with a Lachat QuikChem 8500 flow injection analyzer (Lachat Instruments, Milwaukee, WI, USA) [45]. AP was extracted using 0.5 M sodium bicarbonate and measured using the molybdenum blue spectrophotometric method [46]. TN was determined using the Kjeldahl digestion method [47], while SOC was assessed using the K₂Cr₂O₇ wet oxidation method [48]. AK, NH₄⁺-N, solid-sucrase (S-SC), solid-β-glucosidase (S-β-GC), solid-urease (URE), and soil alkaline phosphatase (ALP) were measured in 0.2 g of dried soil (passed through a 50-mesh sieve), 0.1 g of fresh soil (passed through a 40-mesh sieve), 0.1 g of fresh soil, 0.05 g of dried soil, 0.1 g of dried soil, and 0.05 g of dried soil, respectively, with the relevant kits provided by Guangzhou Redshine Biotechnology Co., Ltd., Guangzhou, China.

2.4. Amplicon Sequencing and Soil Microbial Community Analysis

DNA extraction and high-throughput sequencing were performed by Novogene Biotechnology Co., Ltd. in Beijing, China (https://www.novogene.com, accessed on 10 March 2023). The CTAB method was utilized to extract genomic DNA from samples. The purity and concentration of the DNA were evaluated using agarose gel electrophoresis. A suitable volume of the sample was transferred into a centrifuge tube and diluted to 1 ng/μL using sterile water. Using the diluted genomic DNA as a template, PCR was conducted with barcode-tagged 16S rRNA-specific primers (27F: AGRGTTTGATYNTGGCTCAG, 1492R: TASGGHTACCTTGTTASGACTT) [49]. We used Phusion^® High-Fidelity PCR Master Mix with GC Buffer (New England Biolabs, Massachusetts, USA) and high-fidelity polymerase to ensure optimal amplification efficiency and accuracy. The PCR products were analyzed by electrophoresis using a 2% agarose gel. Equal amounts of PCR products were pooled based on their concentration, thoroughly mixed, and reanalyzed by 2% agarose gel electrophoresis. The target bands were subsequently extracted using the QIAquick^® Gel Extraction Kit (QIAGEN, Hilden, Germany). DNA ligase was employed to attach sequencing adapters to both ends of the amplified DNA fragments, and AMPure PB magnetic beads (Pacific Biosciences, Menlo Park, CA, USA) were used for purification, resulting in the construction of an SMRT Bell library. The purified fragments were re-solubilized in a buffer, and specific-sized fragments were selected using BluePippin (Sage Science, Hercules, CA, USA), followed by further purification of the DNA fragments with AMPure PB magnetic beads. The library was quantified using a Qubit, and the size of the inserted fragments was assessed with an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Sequencing was subsequently performed on the PacBio platform. The PacBio output data were exported in BAM format, and Lima software (version 2.0.1) was utilized to differentiate data from various samples based on barcode sequences.

We followed the “DADA2 + PacBio” workflow [50], which is available at https://benjjneb.github.io/LRASManuscript/LRASms_fecal.html (accessed on 4 July 2024); thus, sequencing results (raw data) were processed using DADA2 [50,51] to obtain amplicon sequence variants (ASVs). Subsequently, the 16S rRNA genes were classified and annotated using the SILVA database (release_138) [52]. ASVs with total sequence counts of fewer than 10 and frequencies of less than 3 were removed to eliminate potential sequencing errors.

The statistical analysis and visualization of microbiome data were conducted in R (version 4.3.0, https://www.r-project.org/). The results of the bioinformatics analysis were imported into R, where the cal_alphadiv function within the microtable class of the microeco package [53] was used to estimate α-diversity based on the Shannon index. β-diversity was evaluated using the cal_betadiv function based on the Bray–Curtis index. To test the effect of treatments on β-diversity, the trans_beta class in the package was employed to perform Permutational Multivariate Analysis of Variance (PerMANOVA) with 999 permutations. Prokaryotic functional traits were predicted using the FAPROTAX database [54] based on 16S feature information through the trans_func class in the microeco package. The trans_venn class in the microeco package was utilized to analyze cultivar-specific microorganisms. To compare the impacts of different cultivars on soil physicochemical properties, ANOVA was conducted using the trans_diff class in the microeco package. The metabolic functions of soil bacteria associated with different cultivars were predicted based on 16S rRNA sequences using the PICRUST2 database [55]. The Mantel test was performed using the linkET package (http://github.com/Hy4m/linkET, accessed on 16 October 2024) to explore the relationships between microbial community differences and their functional traits.

2.5. Analysis of Microbial Community Complexity and Stability

Microbial network analysis was conducted using the Meconetcomp package [56]. The Benjamini and Hochberg method was employed to adjust the false discovery rate (FDR) of p-values. The results were filtered using thresholds of correlation coefficients (r) > 0.6 and significance (p) < 0.01, while values with relative abundance < 0.0001 were excluded. Network topology parameters were extracted using the Gephi platform (version 0.10.0, http://gephi.github.io/). The complexity of the network was primarily assessed by calculating standardized scores for the extracted network topology parameters: Positive Cohesion, Node, Edge, Eigenvector Centrality, Average Degree, Average Clustering Coefficient, Average Path Length, Connectance, and Modularity. An integrated index reflecting microbial network complexity, referred to as the network complexity index, was established by averaging these standardized scores, calculated using the following formula [57]:

$${\widehat{\mathrm{X}}}_{\mathrm{ij}}={\displaystyle \frac{\left({\mathrm{X}}_{\mathrm{ij}}-{\mathrm{X}}_{\mathrm{jmin}}\right)}{\left({\mathrm{X}}_{\mathrm{jmax}}-{\mathrm{X}}_{\mathrm{jmin}}\right)}}$$

(2)

where X_raw, X_min, and X_max represent the raw topological attributes, the minimum value, and the maximum value, respectively; Positive Cohesion is a method used for quantifying the connectivity of a microbial community [58].

Robustness refers to a network’s resistance to node loss, which is evaluated by randomly removing a specific proportion of nodes and measuring the network’s stability [59].

2.6. Construction of Minimum Dataset and Soil Quality Assessment

In this study, 15 parameters including the physicochemical properties of the bulk soil and enzyme activity (SWC, TN, NO₃⁻-N, NH₄⁺-N, AK, AP, pH, ALP, URE, S-SC, S-β-GC, and SOC), together with three soil microbial community indexes (the Shannon diversity index, which considers both species richness and evenness, being more sensitive to rare species and thus providing a more comprehensive assessment of soil microbial diversity than the Simpson index; stability; and complexity), were used to create a comprehensive dataset. In R (version 4.3.0), the scale( ) function was first used to standardize the data, after which the minimum dataset (MDS) index was screened, and the soil quality index (SQI) was calculated [60,61] as follows:

$$\mathrm{SQI}=\sum _{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{W}}_{\mathrm{i}}\mathrm{F}\left(\mathrm{x}\right)$$

(3)

where W_i is the weight of the i-th index and F(x) is the membership value of the i-th index.

The methodology for selecting the MDS indexes and calculating the SQI is based on Qiu et al. [61]. The determination of the membership function parameters for soil physiochemical properties and enzyme activities was based on Chen et al. [60]. According to local soil characteristics and expert experience, all physiochemical properties and enzyme activities were included in the S-shaped membership function, except for pH, which is a parabolic membership function. Additionally, the soil microbial data (Shannon, complexity, and stability) were also incorporated into the S-shaped membership function. To further explore the relationships among the indexes, correlation analysis was conducted using the Corrplot package (https://github.com/taiyun/corrplot, accessed on 17 October 2024).

3. Results

3.1. Physiological and Growth Characteristics and Comprehensive Evaluation, of Different Cultivars

The results of this study (Figure 1) revealed no significant differences among the cultivars in growth characteristics such as plant height, breast diameter, and ground-level diameter (Figure 1A–C). However, we did find differences in physiological characteristics such as root phosphorus (P) and nitrogen (N) concentrations and net photosynthetic rate (Figure 1D–F), although stomatal conductance was statistically identical across all cultivars (Figure 1G). For nutrient absorption efficiency (root P and N concentrations), cultivars 6 and 16 were significantly higher than the other cultivars, while the remaining cultivars could not be distinguished from one another (Figure 1D,E). Notably, cultivar 6, which had the highest root P concentration, significantly exceeded that of cultivar 16, whereas the difference in root N concentration between cultivars 6 and 16 was not significant. Additionally, for the net photosynthetic rate (Figure 1F), cultivar 6 had a significantly higher rate than cultivar 4, with the other cultivars showing no significant differences. A comprehensive evaluation of growth using the membership function (Figure 1H) also indicated differences in performance among the cultivars, with cultivar 6 achieving the highest comprehensive score, significantly surpassing the others, while cultivar 4 had the lowest score, and the remaining cultivars were comparable.

3.2. Physicochemical Properties and Enzyme Activity of Bulk Soil in Different Cultivars

The soil samples associated with all cultivars had a significantly higher moisture content than the control, CK (Figure 2A), with cultivars 3 and 5 showing the highest levels and cultivars 6 and 8 the lowest; the other cultivars showed no significant differences. Regarding soil pH (Figure 2B), cultivars 3 and 8 were comparable to CK (pH 4.4), while the other cultivars (pH 4.5–4.8) were significantly higher. For available soil nutrients (Figure 2C–F), all cultivars were significantly higher than or equivalent to CK, and there were some differences among the cultivars. Specifically, for AP and AK, all cultivars were significantly higher than CK. Cultivar 8 had the highest soil AP content, while cultivar 16 had the lowest, with the other cultivars being comparable (Figure 2C). For soil AK, cultivars 4 and 8 had the highest levels, while cultivar 5 had the lowest, with the others falling in between (Figure 2D). In terms of NO₃⁻-N, cultivars 5 and 8 had significantly higher levels than the other cultivars, which were comparable to CK (Figure 2E). For NH₄⁺-N, cultivars 4, 5, 6, and 8 were significantly higher than CK (Figure 2F), while the remaining cultivars were comparable to the control. Similarly, in terms of soil nutrients (Figure 2G, H), the cultivars were also higher or comparable to CK. For TN, cultivars 3, 5, and 16 were significantly higher than CK, and the other cultivars were comparable to CK (Figure 2G). For SOC, cultivars 4 and 8 had the highest values, followed by cultivar 6, all of which were significantly higher than CK, with the other cultivars showing no significant differences (Figure 2H). In terms of enzyme activity, all cultivars were significantly higher than CK (Figure 2I–L). For S-SC activity, cultivars 6 and 8 exhibited the highest levels, while cultivars 3 and 5 showed the lowest (Figure 2I). For S-β-GC and URE activity, there were no significant differences among cultivars (Figure 2J,K), while for ALP activity, cultivars 3, 5, and 16 were similar but significantly higher than cultivars 4, 6, and 8, which were comparable to each other (Figure 2L).

3.3. Microbial Diversity and Species Composition in Bulk Soils of Different Cultivars

The α diversity (Shannon index) of soils associated with the various cultivars and CK is shown in Figure 3A. Except for cultivars 8 and 16, which were significantly higher than CK, there was no significant difference compared to the control. Principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarity (Figure 3B) indicated significant differences in β diversity among the bulk soils of different cultivars and CK (Adonis R² = 0.22, p = 0.003). Although there were differences in the biodiversity of the soil associated with the various cultivars of F. falcata and the control, the composition of the respective microbial communities was very similar at the phylum level (Figure 3C). Proteobacteria was the most dominant phylum across all cultivars and CK, accounting for 34.1% to 76.2% of the total community, with the highest relative abundance observed in cultivar 6 (76.2%). Cultivars 8 (45.8%) and 16 (34.1%) had lower abundances than CK (56.9%), while the relative abundances of the other cultivars were comparable to CK. Acidobacteriota was the second most dominant phylum, representing 9.6% to 28.4% of the total community, with cultivars 8 (20.7%) and 16 (28.4%) showing significantly higher relative abundances than CK (17.6%). In contrast, Acidobacteriota was represented at a significantly lower level in soil associated with cultivar 6 (9.6%) than in CK, while the relative abundances in cultivars 3, 4, and 5 (12.7% to 15.1%) were similar to the control. At the genus level (Figure 3D), all cultivars except cultivar 16 exhibited a composition similar to that of CK. Apart from cultivar 6, Pseudomonas was the major dominant genus, with its representation ranging from 18.3% to 65.6% and the highest relative abundance being in cultivar 6. Soil associated with cultivar 8 (18.3%) showed the lowest representation of Pseudomonas, even lower than CK (37%). However, cultivar 16 (0.1%) almost completely lacked Pseudomonas, while its other major genera were largely consistent with those of the other cultivars.

3.4. Complexity and Stability of Bulk Soil Microbial Communities in Different Cultivars

A comprehensive evaluation of the network topology properties of bulk soil microbial communities across different cultivars revealed that, compared with CK (0.845), all cultivars exhibited a notable reduction in microbial community complexity (

Table 1. Network topological properties, complexity, and stability of soil microbial community across different cultivars of Falcataria falcata (L.) Greuter & R. Rankin.

Group	PC	Node	Edge	EC	AD	ACC	APL	C	Q	Complexity	Stability
CK	0.894	1282	43239	0.291	67.456	0.972	1.049	0.053	0.848	0.845	4.300
Cv. 3	0.804	1007	21451	0.097	42.604	0.956	1.094	0.042	0.761	0.526	4.480
Cv. 4	0.789	824	9106	0.080	22.102	0.956	1.038	0.027	0.804	0.360	0.360
Cv. 5	0.898	951	19575	0.134	34.858	0.944	1.273	0.037	0.815	0.571	1.530
Cv. 6	0.858	629	9430	0.014	29.984	0.976	1.007	0.048	0.534	0.342	5.180
Cv. 8	0.514	1456	17431	0.218	23.944	0.891	1.369	0.016	0.928	0.492	0.020
Cv. 16	0.382	1363	11339	0.012	16.638	0.944	1.052	0.012	0.950	0.300	0.000

PC, EC, AD, ACC, APL, C, and Q represent Positive Cohesion, Eigenvector Centrality, Average Degree, Average Clustering Coefficient, Average Path Length, Connectance, and Modularity, respectively. Stability reflects the robustness of the network and is quantified by the absolute value of the slope of the regression equation for each group.

). Cultivars 4, 6, 8, and 16 showed the most significant decreases in complexity, ranging from 0.300 to 0.492, while cultivars 3 and 5 experienced relatively smaller reductions (0.526 to 0.571). Network stability was evaluated by the effect of node removal on natural network connectivity (Table 1). The results showed that cultivars 3 and 6, along with CK, exhibited a rapid decline in natural connectivity, suggesting lower network stability (4.300 to 5.180). In contrast, other cultivars displayed slower rates of decline in natural connectivity, indicating higher network stability (0 to 1.530), with cultivars 8 and 16 being the most stable (0.000 to 0.020).

3.5. Specific Microbes and Functional Characteristics of Bulk Soil in Different Cultivars

We investigated the number of unique microbes in the bulk soil associated with different cultivars and show these in a petal map (Figure 4A), which reveals specific differences in the composition of the microbial communities. Cultivar 16 recruited the most unique microbes (241), followed by cultivar 8 (90), while the other cultivars had relatively fewer recruits (12–31), with cultivars 4 and 6 having the fewest (12–13). Predictions based on the FAPROTAX database (Figure 4C) indicate that these unique members of the microbial communities exhibit differences in biogeochemical cycling functions. In the carbon cycle, cultivars 4 and 5 primarily recruited unique microbes related to anaerobic chemoheterotrophy, with cultivar 5 showing strong performances in fermentation and cellulose decomposition. Cultivars 6, 8, and 16 mainly recruited unique microbes associated with aerobic chemoheterotrophy, while the unique recruits associated with cultivar 3 demonstrated weak performance in the carbon cycle. In the nitrogen cycle, cultivar 8 primarily recruited unique microbes related to biological nitrogen fixation, while those associated with cultivar 16 were involved with nitrate reduction; the other cultivars showed a weak performance in this area. For the iron cycle, only cultivar 16 showed significant recruitment, with a number of microbes related to iron respiration. The metabolic functions of the microbial communities in different cultivars and CK were also predicted by the PICRUSt2 database, identifying a total of 6494 Kyoto Encyclopedia of Genes and Genomes orthologs (KOs). The Mantel test was used to evaluate the relationship between the structure of the microbial communities and their functional characteristics (Figure 4B), revealing significant co-variation between the metabolic functions and structures of the microbial communities in the bulk soils of different cultivars and CK (R² = 0.64, p < 0.001).

3.6. Soil Quality Assessment

Principal component analysis (PCA) was performed on 15 parameters of the bulk soil physicochemical properties, enzyme activities, and microbial communities across the various cultivars. The results (Table S1) indicate that there are three principal components with eigenvalues ≥ 1, which cumulatively account for 62.10%, demonstrating good explanatory capacity for these components. All indexes were filtered, with those having absolute loadings ≥ 0.5 being grouped together. If the same index was present in two principal components, it was included in the group with the lower correlation (Table S1). The first group included AP, AK, SOC, (S-SC, S-β-GC, URE, ALP, and complexity; the second group included SWC, NH₄⁺-N, and TN; the third group included pH, the Shannon index, and stability. Then, indexes within a 10% range of the highest Norm value for each group were retained. The indexes retained from the first group were S-SC, with a Norm value of 2.204; the second group retained SWC, NH₄⁺-N, and TN, with Norm values of 1.547, 1.435, and 1.396, respectively; and the third group retained stability, with a Norm value of 1.307. If two indexes within the same group showed a high correlation (correlation coefficient ≥ 0.5), the index with a higher Norm value was retained. Ultimately, the four indexes (of the original 15) that were included in the minimum dataset (MDS) were S-SC, SWC, NH₄⁺-N, and stability (Table S2), achieving a filtering rate of 73.3%. The weight value of each index was calculated (Table S2). Correlation analysis was subsequently performed on all indexes using the Corrplot package (Figure 5A). PCA was conducted to obtain the common factor variance of the total dataset and the minimum dataset indexes, and the weight values for each index were then calculated (Table S2). Finally, the SQI for the total dataset (TDS-SQI) and the SQI for the minimum dataset (MDS-SQI) were calculated, and a linear fit was performed between the two. The results (Figure 5B) indicate a significant linear relationship between MDS-SQI and TDS-SQI (R² = 0.783, p < 0.001), suggesting that the soil quality index calculated based on the MDS method can be used to assess the bulk soil quality of each cultivar. The results of a comprehensive assessment of soil quality (Figure 5C) showed that, compared to CK (0.150), all cultivars significantly improved soil quality (0.471–0.722), with significant differences among the cultivars. Cultivars 8 and 16 were the best, significantly higher than cultivars 3, 4, and 6 while showing no significant difference from cultivar 5.

4. Discussion

Leguminous plants enhance the nitrogen content of soil through nitrogen fixation, significantly improving soil fertility [62,63]. Their root systems and the decomposition of residual plant materials further enhance the physical and chemical properties of the soil [64,65], leading to improved soil structure and fertility levels. Consequently, the introduction of leguminous plants is widely recognized for its positive impact on soil quality across various soil types and climatic conditions [66,67]. Current research primarily focuses on the effect of leguminous plants, especially herbaceous legumes, on the physicochemical properties and fertility of soil [68,69,70,71,72,73]. However, studies on how woody leguminous plants (particularly fast-growing economic species) affect soil quality and the microbial communities in bulk soil, and how these are related, remain limited [74,75]. Numerous studies have demonstrated that soil microbial communities are essential for maintaining soil ecosystem services [76,77]. Therefore, it is crucial to investigate the effects of woody leguminous species on soil fertility and microbial communities and explore the relationship between these communities and soil quality. The results of this study indicate that, in the early stages of cultivation, different cultivars of F. falcata can significantly improve soil quality, but there are variations among cultivars in physiological and growth characteristics, soil physicochemical properties, and the structure and function of soil microbial communities. This suggests that the various cultivars of F. falcata may contribute to soil fertility through different mechanisms.

4.1. Different Cultivars of F. falcata Can Promote Soil Fertility by Recruiting Specific Functional Microorganisms During the Early Stages of Cultivation

Previous studies [78] have shown that different varieties of the same plant can have significantly different effects on soil nutrients. This variation primarily stems from differences in nutrient absorption capabilities and the quantity and quality of organic compounds secreted by the roots of each variety [79,80]. The results of our study also indicate varying degrees of difference in root P and N concentrations, as well as in the physicochemical properties, nutrient levels, and enzyme activities of the bulk soil associated with different cultivars of F. falcata. Additionally, PCoA revealed significant differences in the composition and structure of soil microbial communities (β diversity) among the various cultivars. This suggests that different cultivars exert a significant selective effect on soil microbial communities during the early stages of planting; that is, each cultivar recruits specific microorganisms to the bulk soil associated with it. Previous research has indicated that crop varieties may be key drivers of soil bacterial community composition [81,82,83]. The functional prediction results based on the FAPROTAX database showed that the specific microorganisms recruited by different cultivars of F. falcata exhibit unique biogeochemical cycling functions. Moreover, the results from the Mantel test indicate a significant co-variation between the metabolic functions of soil microbial communities and their community structures in the bulk soils of different cultivars. For example, cultivars 8 and 16 primarily recruited microorganisms associated with nitrogen fixation and nitrate reduction, respectively. The levels of available NO₃⁻-N and NH₄⁺-N in their bulk soils were significantly higher than those in the control (CK) and in other cultivars. This phenomenon may be attributed to the crucial role of these microorganisms in the soil nitrogen cycle. Nitrogen-fixing microorganisms can convert atmospheric N₂ into plant-available NH₄⁺ through biological nitrogen fixation, thereby increasing ammonium nitrogen content [84,85]. Meanwhile, microorganisms associated with nitrate reduction may regulate the dynamic balance of soil NO₃⁻ through denitrification or dissimilatory nitrate reduction, thereby influencing the availability of nitrate nitrogen. The synergistic enhancement of these processes may be the primary driving mechanism behind the significant increase in NH₄⁺-N and NO₃⁻-N content in the soil of cultivars 8 and 16. This indicates that cultivars 8 and 16 contribute significantly to the improvement of soil nitrogen fertility. Cultivars 6, 8, and 16 recruited specific microorganisms related to aerobic chemoheterotrophic carbon cycling, with S-SC enzyme activity levels in their bulk soils significantly surpassing those in CK and other cultivars. This phenomenon may be attributed to the ability of these microorganisms to enhance the decomposition rate of soil organic matter, thereby promoting more efficient carbon and nutrient cycling [86,87], which in turn significantly increased soil S-SC activity levels. This suggests that these three cultivars can more effectively enhance the decomposition of organic matter in soil and thus the ability to release nutrients [88]. Overall, these findings indicate that different cultivars of F. falcata can selectively recruit functionally different microorganisms, thereby influencing soil physicochemical properties, nutrient availability, and enzyme activities. This may have complex and far-reaching effects on long-term ecological processes in soil. The specificity of these interactions may be closely linked to differences in root exudates among cultivars, as the secretion characteristics of roots from different cultivars can shape not only the structure and function of rhizosphere microbial communities but also further impact soil structure and stability [89,90].

It is noteworthy that while key soil factors that influence the composition and structure of soil microbial communities have been extensively studied and confirmed across various ecosystems [91,92,93], the mechanisms by which different crop varieties, particularly leguminous plants such as F. falcata, recruit specific microorganisms remain unclear. This area warrants further in-depth and systematic research in the future.

4.2. The Stability of the Soil Microbial Community Is the Key Biological and Ecological Index for the Evaluation of Quality in Bulk Soil Associated with F. falcata

The quality and health of the soil, as a complex and dynamic ecosystem, are closely dependent on the soil microbiome. Studies have shown that soils with higher microbial diversity often exhibit improved ecological functions, resilience to environmental stress, and enhanced crop productivity [8,94,95,96]. Current evaluations of soil quality and health typically include physical, chemical, and biological indexes. However, physical and chemical indexes alone cannot adequately reflect dynamic changes in soil as a living system. Thus, soil quality is primarily related to the microbial components and ecological functions of the soil ecosystem, particularly those that maintain energy flow, material cycling, and information exchange [8,10,13]. As a result, biological indexes of soil, such as microbial composition, structure, function, and biological processes, have garnered increasing attention from researchers in recent years [8,97,98]. Due to the complexity of soil ecosystems, our understanding of the relationship between soil microbial communities and soil health remains limited [8]. Research indicates that among various biological indexes, the composition of the soil microbiome is an important indicator of soil health [7,99]. For instance, the Shannon index (Biodiversity), which measures species diversity by accounting for the relative abundance of species, has been shown to correlate closely with soil productivity. Studies have shown that a reduction in microbial diversity (Shannon index) will significantly reduce plant biomass, underscoring the crucial role of soil microbial diversity in maintaining ecosystem functions [100]. Higher community network complexity indicates tighter connections among nodes [101], and the intricate interconnections between microorganisms may help sustain ecosystem functions. Consequently, more diverse and complex microbial networks can enhance multiple functions due to their greater functional redundancy and the presence of unique species [102]. Additionally, the stability of microbial communities (networks) is essential for maintaining the sustainability of soil ecosystem services and serves as a key measure of soil ecosystem health. Therefore, this study assessed three biological indexes representative of soil ecosystem characteristics, namely biodiversity (measured by the Shannon index), microbial community complexity, and stability, alongside physicochemical indexes, to comprehensively evaluate soil quality using a minimum dataset (MDS) approach. The results indicate that the bulk soils associated with six cultivars of F. falcata significantly outperformed the control samples in quality, with notable differences among the varieties. Cultivars 8 and 16 are identified as optimal, significantly outpacing the others. However, when using traditional methods that rely solely on physicochemical indexes and enzyme activity (Figure S1), the bulk soils of all six cultivars score more highly than controls, with no significant differences observed among the cultivars. This suggests that without incorporating microbial indexes, only the physicochemical and fertility status of the soil is reflected, failing to capture the status of microbial communities and ecological function and thus not accurately representing actual soil health. In fact, soil microbial communities play a key role in element cycling [103,104], soil degradation [105], enhancing rhizosphere immunity [106], and improving soil fertility and crop yield [13,107]. Through the proper management and utilization of soil microorganisms, significant improvements in soil health and agricultural productivity can be achieved, positively impacting human health and environmental protection [8]. Therefore, integrating key soil ecological and biological indexes with physicochemical indexes provides a more accurate reflection of the actual health status of soil.

This study identified four indexes for the MDS derived from 15 soil physicochemical and microbial indexes, i.e., S-SC, SWC, NH₄⁺-N, and stability, which most effectively reflect the health status of F. falcata bulk soils and are key factors influencing their quality. Notably, the only microbial index included is microbial community stability, which is crucial since the stability of soil microbial communities is often determined by biodiversity, species composition, structure, and metabolic functions. Thus, the stability of a microbial community is closely linked to the functions and services of the soil ecosystem, providing a comprehensive reflection of soil health. In this study, cultivars 8 and 16, which had the best bulk soil quality, also showed the highest stability and biodiversity (Shannon) in their respective microbial communities (Figure 3A). In contrast, the cultivars with poorer soil quality (3, 4, and 6) showed lower microbial community stability and biodiversity. This indicates a positive correlation between the stability of soil microbial communities and biodiversity. Furthermore, our results demonstrate that cultivars 8 and 16 recruited the highest number of unique microorganisms with counts of 241 and 90, respectively, significantly exceeding those found in other cultivars (12–31). This suggests a positive correlation between the stability of microbial communities and the number of unique microorganisms recruited. An intriguing finding was that, in terms of species composition, the relative abundance of Proteobacteria (Pseudomonas) in cultivars 8 and 16 was notably lower than in other cultivars and CK. This phenomenon may be closely related to microbial adaptation strategies in response to ecosystem changes. The adaptation strategies of microorganisms evolve during the evolution of the soil ecosystem alongside diversification and changes in the dominant components of the microbial community, which are, in turn, responsible for certain soil biological processes [108]. Specifically, in the early stages of cultivation, cultivars 8 and 16 may have facilitated the rapid succession or transformation of the soil ecosystem through root exudates or genetic traits. During this process, the community structure exhibited a trend in which fast-growing, copiotrophic microorganisms represented by Proteobacteria were gradually replaced by microbial populations better adapted to oligotrophic environments. This dynamic shift may be one of the reasons for the higher stability of the soil microbial communities associated with cultivars 8 and 16. Overall, these findings suggest that the stability of soil microbial communities is a critical biological and ecological factor influencing the quality and health of F. falcata bulk soils. It serves as a key biological index for assessing soil health, providing valuable references for evaluating the status of soil quality and health in other plants (crops).

4.3. Selection Strategies for F. falcata Cultivars in Forestry Cultivation Practices

This study found significant differences among various cultivars of F. falcata in key physiological and growth characteristics, soil quality, and soil microbial composition during the early stages of cultivation. The results provide important theoretical support for cultivation practices and breeding strategies. While the basic growth traits of plants, such as height, diameter at breast height, and ground-level diameter, are largely influenced by external environmental factors and may not show significant differences initially, inherent genetic variation and physiological characteristics may play crucial roles in determining plant growth potential [109]. Therefore, when selecting cultivars (varieties), not only should early morphological traits be considered but also physiological characteristics, including nutrient absorption efficiency and photosynthetic capacity. Our results indicate that although no significant differences in growth traits were found among the cultivars during the early stages of cultivation, there were notable differences in nutrient absorption efficiency and photosynthetic capacity. Cultivars 6 and 16 demonstrated outstanding nutrient absorption efficiency, particularly for phosphorus (P) and nitrogen (N), with root nutrient concentrations significantly higher than those of other cultivars. This suggests that these two cultivars can utilize soil nutrients more effectively, supporting rapid early growth [110]. Photosynthetic capacity is also vital for growth potential [111,112]. We found that all cultivars except cultivar 4 exhibited strong photosynthetic abilities, with cultivar 6 performing the best, while differences among the other cultivars were not significant. A comprehensive evaluation of physiological and growth traits gave cultivar 6 the highest overall score, significantly surpassing the other cultivars, while cultivar 4 scored lowest, with no significant differences among the remaining cultivars. This suggests that cultivar 6 is optimal during the early stages of cultivation, while cultivar 4 is the least favorable, with the others showing no significant differences. Another critical factor affecting healthy plant growth is soil quality, which is closely related to the composition of the soil microbial community. Therefore, the selection of planting varieties (cultivars) should also consider their impact on soil quality. A comprehensive evaluation of soil quality revealed that cultivars 8 and 16 scored best, having the highest microbial community stability, biodiversity (Shannon index), and unique microbial counts. Additionally, their bulk soils had significantly higher levels of NO₃⁻-N and NH₄⁺-N, as well as S-SC activity, compared to other cultivars, indicating a more robust and resilient bulk soil ecosystem, which is advantageous for improving soil fertility. Thus, from the perspective of soil fertility, quality, and health, cultivars 8 and 16 are the most favorable, followed by cultivar 5, while cultivars 3, 4, and 6 are relatively less useful.

In summary, cultivar 6 excels in nutrient absorption efficiency and photosynthetic capacity, making it a recommended cultivar for economic forestry aimed at timber production. Conversely, cultivars 8 and 16 significantly outperform other cultivars in their effect on soil quality and health, as well as ecological stability, although their overall physiological and growth evaluation is lower than that of cultivar 6. Therefore, these two cultivars are excellent candidates for long-term cultivation in public ecologically important forests, contributing to the maintenance of soil quality and health and the sustainability of soil ecosystem services.

5. Conclusions

This study integrated soil microbial community indexes with soil physicochemical properties to calculate a soil quality index (SQI) for evaluating the impact of different cultivars of F. falcata on soil quality and health. This is a new approach to addressing soil degradation, improving soil quality, and restoring ecosystem services. The results show that, in the early stages of cultivation, different cultivars significantly enhance soil fertility by recruiting specific functional microorganisms. Notably, the stability of the soil microbial community emerged as a critical biological and ecological factor that influences and maintains the quality and health of the bulk soil, serving as an essential index for the assessment of soil quality and health. Cultivar 6 demonstrated superior growth properties, making it suitable for economic forestry plantations. Conversely, cultivars 8 and 16 excelled in improving soil quality and health, showcasing significant potential to enhance soil ecosystem services, and they are therefore recommended for planting in ecologically important forests.