Short-Term Effects of Three Tree Species on Soil Physicochemical Properties and Microbial Communities During Land-Use Change from Farmland to Forests

Abstract

In recent decades, much of China’s farmland has been transformed into forests due to the Conversion of Farmland to Forests and Grasses Project. While past research has mainly examined soil nutrients and water conservation, less attention has been given to soil microbial communities. This study examined the effects of converting farmland to forests of Pleioblastus amarus (PA), Populus deltoides (PD), or Zanthoxylum bungeanum (ZB) on the soil physiochemical properties, enzymes, and microbial communities, using abandoned land (AL) as the control, over a period of five years. The results showed that PA increased the soil organic carbon (SOC) content, although not significantly, while significantly boosting the C:N and C:P ratios and urease activity compared to the AL. PD notably reduced the amylase and cellulase activities, as well as the fungal Shannon index. Additionally, the beta diversity of both the bacterial and fungal communities in the PA stand was clearly distinct from that of the AL and the other tree species. The SOC content, total potassium content, and cellulase activity showed significant correlations with bacterial communities. Moreover, the bacterial community changes in the PD and ZB stands were mainly driven by the genera Steroidobacter, Roseisolibacter, and Serendipita, and were negatively correlated with the SOC content, C:N and C:P ratios, and cellulase activity. In contrast, the fungal community changes in the PA stand were primarily influenced by the order Capnodiales, family Capnodiaceae, genus Chaetocapnodium, and species Chaetocapnodium philippinense, which were positively correlated with the soil pH, C:N and C:P ratios, and cellulase activity. Furthermore, “Metabolism” was identified as the primary bacterial function, and converting farmland to forest altered the fungal nutritional type from Saprotroph to Pathotroph–Saprotroph–Symbiotroph, particularly in the PA stand. These findings indicate that converting farmland to forest, particularly with bamboo P. amarus, significantly impacts the bacterial and fungal communities in the soil and changes the fungal trophic type due to the carbon source and cellulase activity of this tree species.

1. Introduction

In recent decades, due to the Returning Farmland to Forest Program (RFFP)—also known as the “Conversion of Farmland to Forests and Grasses Project” or the “Grain for Green Program”—alongside urbanization and the necessity for ecological security, a substantial portion of farmland in China has been transformed into forests or abandoned land [1,2]. From the initiation of the first phase of the RFFP at the close of the 20th century to 2020, over 14.23 million hectares of farmland and 17.53 million hectares of barren hills and abandoned land were transformed into forests [2]. Furthermore, 5.85% of the cultivated land in China’s primary grain-producing regions have been abandoned, primarily due to the migration of farmers to urban areas, driven by the rapid advancement of industrialization and urbanization [3]. It is important to highlight that the transformation of farmland into afforested areas is not solely a reaction to urban expansion but also a deliberate strategy to tackle the critical challenges of environmental degradation and carbon emissions [4,5]. Accumulating evidence has demonstrated that the RFFP has significantly mitigated the issues related to soil erosion and frequent river flooding, while generally enhancing forest coverage and improving household livelihoods in China since 1999 [6,7].

In addition to the ecological benefits, numerous studies have indicated that the conversion of farmland to forests significantly impacts the soil’s physicochemical properties, with considerable variation observed depending on the tree species [8,9]. Nevertheless, the findings were not consistent and were affected by multiple factors. For example, in a subtropical karst region of southwest China, the conversion of farmland to forest resulted in significant increases in the soil organic carbon (SOC) and total nitrogen (TN) levels [10]. Furthermore, a study carried out in Northeast China demonstrated that the afforestation of agricultural land resulted in significant increases in the SOC and TN levels, as well as increases in soil pH and decreases in bulk density [11]. Additionally, various vegetation restoration techniques have been found to enhance soil quality to differing extents, with native tree species frequently producing more favorable outcomes compared to non-native species [12]. Moreover, the choice of tree species, such as Caragana korshinskii and Robinia pseudoacacia, can lead to substantial variations in soil quality parameters, including the SOC, TN, total phosphorus (TP) levels, and soil phosphatase activity [13]. Therefore, a substantial body of research is required to thoroughly elucidate the effects on soil properties across different study areas or using different afforested tree species.

Soil microbial communities constitute critical biotic elements within terrestrial ecosystems, significantly contributing to the translocation of matter and energy and the regulation of nutrient cycles [14]. Recent research has investigated the impact of soil properties, including pH, nutrient content, and enzymatic activity, on these microbial communities [15]. However, the findings of previous studies have remained inconclusive. For example, soil pH has been demonstrated to substantially affect the composition and diversity of microbial communities, with some studies suggesting that bacterial communities exhibit greater sensitivity to pH fluctuations compared to fungal communities [16]. In a controlled laboratory experiment, it was determined that the diversity and composition of bacterial communities were strongly correlated with soil pH, whereas the fungal diversity appeared to be less influenced by pH variations [17]. Furthermore, the interaction between soil pH and microbial community composition has been demonstrated to affect enzyme activities, with particular bacterial groups showing distinct responses to alterations in soil chemistry [18]. This suggests that the relationship between soil enzyme activities and microbial communities is intricate and context-dependent, exhibiting variability across different soil types and environmental conditions. Consequently, extensive research is required to elucidate the relationship between soil physicochemical properties—particularly those influenced by various afforested tree species—and microbial communities.

Different tree species markedly alter soil properties, which could subsequently affect microbial communities and their associated functional profiles or nutritional types during the initial stages of forest development following the conversion of farmland to forest (five years post-afforestation). To test this hypothesis, we assessed the soil physical, chemical, and enzymatic properties in three representative woody plantations—containing a bamboo species, a fast-growing timber species, and an economically valuable species—and compared these findings to those from an abandoned land area, which served as a control. Simultaneously, 16S rRNA and ITS high-throughput sequencing technologies were utilized to assess the diversity and composition of the bacterial and fungal communities in the soil, respectively. The objectives of the study were to (i) investigate the effects of afforestation on soil properties and microbial communities, (ii) evaluate the response of soil microbial communities to alterations in these properties, and (iii) the effect of the afforested tree species on microbial functions.

2. Materials and Methods

2.1. Site Description

The experiment was conducted at the modern agricultural research and development base of Sichuan Agricultural University, Chongzhou, Chengdu, Sichuan Province, China (103°39′55″ E and 30°32′54″ N; altitude: 520.6 m). The location is characterized by a subtropical monsoon climate, with an annual mean temperature of 17.1 °C. The recorded temperature extremes include a high of 36.6 °C in August and a low of −4.2 °C in January. The annual precipitation ranges from 802.5 mm to 1343.3 mm, and the average relative humidity is 75%, fluctuating between 44% and 91%. The region receives an average of 1156.3 h of sunshine per year, and the cumulative annual temperature above 0 °C totals 6262 °C. Prior to the afforestation, this site was farmland that had undergone more than 30 years of paddy–upland rotations, with the top 0–20 cm plough horizon extensively tilled [19]. The aquatic crop cultivated was rice, whereas the upland crops comprised wheat, rapeseed, and potatoes. The background soil properties were characterized by a pH of 6.99, soil organic carbon (SOC) content of 18.84 g kg⁻¹, total nitrogen (TN) content of 1.92 g kg⁻¹, total phosphorus (TP) content of 0.58 g kg⁻¹, total potassium (TK) content of 16.84 g kg⁻¹, alkali-hvdrolyzable nitrogen content of 63 mg kg⁻¹, available phosphorus content of 10.43 mg kg⁻¹, and available potassium content of 67.54 mg kg⁻¹. According to the World Reference Base for Soil Resources, the soil was classified as Hortic Anthrosols (Eutric, Loamic, Leptic) [20].

In 2019, three woody plantations containing Pleioblastus amarus (a bamboo species; PA), Populus deltoides (a fast-growing timber species; PD), and Zanthoxylum bungeanum (an economic tree species; ZB) were established using one-year old saplings on a farmland in the modern agricultural research and development base. Each plantation was approximately 1000 m², with a minimum distance of 10 m and a maximum distance of 150 m from each other. Another farmland adjacent to these three plantations with a similar area was left abandoned. In the first three years, these plantations and abandoned land (AL) underwent two rounds of manual loosening and weeding each year using tools such as hoes. This process affected the top layer of approximately 0–3 cm of soil. In the subsequent years, no further forest management was conducted. The accumulated leaf litter were left in place within the stand, and no additional interventions were applied.

2.2. Experimental Design and Soil Sample Collection

In late July 2024, after 5 years of growth, soil samples from the three plantations and the abandoned farmland were undertaken, and the growth state of these four stands are shown in

Table 1. Basic information of the four sample stands.

Tree Species	DBH (cm)	Height (m)	Density (Plant hm−2)	Canopy Density/Cover Density	Dominant Grass Species
AL	—	—	—	0.98	Setaria viridis, Imperata cylindrica, Alternanthera philoxeroides, Bidens pilosa, Pouzolzia zeylanica, Setaria palmifolia, Humulus scandens, Lysimachia christinae, Lactuca indica, Artemisia frigida
PA	4.8	4.5	4230	0.80	Alternanthera philoxeroides, Artemisia frigida, Setaria palmifolia, Imperata cylindrica, Setaria viridis, Chelidonium majus
PD	10.6	11	2500	0.85	Alternanthera philoxeroides, Equisetum hyemale, Artemisia frigida, Symphyotrichum subulatu, Phyllanthus urinaria, Trifolium repens
ZB	4.4	3.7	1667	0.70	Equisetum hyemale, Alternanthera philoxeroides, Trifolium repens, Artemisia argyi, Digitaria sanguinalis

DBH: diameter at breast height; AL: abandoned land; PA: Pleioblastus amarus; PD: Populus deltoides; ZB: Zanthoxylum bungeanum.

and Figure 1. Initially, 4 plots (5 m × 5 m), with a minimum distance of 5 m, were established within each stand. Five soil cores, each measuring 20 cm in height, were extracted from five distinct locations using a soil auger with an inner diameter of 51 mm. This sampling was conducted along an S-shaped sampling line in each plot. Given a short period growth of 5 years after the afforestation treatments, the uppermost diagnostic Ap horizon of the four treatments was identified as the former plough horizon, consisting of a 20 cm deep soil layer [19,20]. The profiles of this horizon exhibited morphological characteristics typical of an Ap horizon, including a diffuse and smooth boundary, brown coloration, a weak fine angular blocky structure, and a consistent thickness across all four treatments. Consequently, the 0–20 cm soil layer was selected for soil sampling for all four treatments, following the removal of surface debris and leaf litter, which were left in situ within the stand. Subsequently, the five soil cores from each plot were mixed to form a composite sample. This composite sample then divided into four portions using the quartering method, and one portion was randomly selected to serve as the final composite soil sample. Consequently, a total of 16 composite soil samples were obtained for this study: one composite soil sample per plot, with four plots per treatment, across four treatments (1 composite soil sample/plot × 4 plots/treatment × 4 treatments = 16 composite soil samples). The composite soil samples were swiftly sieved with a 2 mm mesh sieve to eliminate any rocks and roots. Following sieving, the soil was further split into two portions, with one portion stored in a −80 °C freezer following rapid freezing in liquid nitrogen for future microbial analysis and the remaining portion was kept at 4 °C for immediate assessment of enzyme activities or for analyzing physiochemical characteristics following drying in the open air. Soil core samples from each plot were collected using the ring-knife method to analyze field moisture capacity, total porosity, and soil bulk density.

2.3. Physiochemical Characteristics and Enzymatic Activities

The soil characteristics were assessed following the method of He et al. [21]. To determine the field moisture capacity, the soil was dried at 65 °C until its weight remained unchanged. The ring-knife method [22] was used to measure the total porosity and soil bulk density by weighing the soil and determining the mass after drying the soil in an oven at 105 °C until a stable weight was achieved. The soil pH was assessed using potentiometric titration with a water-to-soil ratio of 1:2.5. The soil organic carbon (SOC) content was quantified through the potassium dichromate oxidation method. The total nitrogen (TN) content was measured using the Kjeldahl method. The total phosphorus (TP) content was determined via molybdenum antimony colorimetry, while the total potassium (TK) content was quantified using alkali-flame photometry. The C:N, C:P, and N:P ratios were determined by dividing the SOC content by the TN content, the SOC content by the TP content, and the TN content by the TP content, respectively.

The soil enzyme activities were evaluated using the methods described by Tie et al. [23]: Acidic phosphatase activity was measured through the spectrophotometric analysis of nitrobenzene phosphate [24]. Sucrase activity was evaluated using the 3,5-dinitrosalicylic acid colorimetric method [25]. Amylase activity was determined using starch as the substrate [26]. Soil cellulase activity was quantified via the 3,5-dinitrosalicylic acid colorimetric method [24]. Lastly, soil urease activity was measured through indigo-phenol colorimetry [27].

2.4. DNA Extraction, Sequencing, and Quantification

The CTAB method was employed to extract genomic DNA from soil samples [28], followed by an assessment of its purity and concentration through 1% agarose gel electrophoresis. Appropriate quantities of DNA were then transferred into a centrifuge tube and diluted to a concentration of 1 ng μL⁻¹ using sterile water. For the amplification of the bacterial 16S rRNA V3-V4 region, primers 341F (5′-CCTAYGGGRBGCASCAG-3′) and 806R (5′-GGACTACNNGGGTATCTAAT-3′) were employed. For the fungal ITS1 region, primers ITS1F (5′-CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2 (5′-GCTGCGTTCTTCATCGATGC-3′) were used. All polymerase chain reaction (PCR) assays were performed using 15 µL of the Phusion High-Fidelity PCR Master Mix (New England Biolabs, Ipswich, MA, USA), supplemented with 0.2 µM of both forward and reverse primers, and approximately 10 ng of template DNA. The thermal cycling protocol commenced with an initial denaturation at 98 °C for 1 min, followed by 30 cycles comprising denaturation at 98 °C for 10 s, annealing at 50 °C for 30 s, and extension at 72 °C for 30 s. A final elongation step was conducted at 72 °C for 5 min. Post-amplification, the PCR products were analyzed using 2% agarose gel electrophoresis. Finally, based on the concentration of the PCR products, an equal amount was mixed and subjected to 2% agarose gel electrophoresis. Following thorough mixing, the bands were cut out and purified utilizing the Qiagen Gel Extraction kit (Qiagen, Hilden, Germany).

Using the NEBNext^® Ultra™ II DNA Library Prep Kit for Illumina (New England Biolabs, Ipswich, MA, USA), a library was constructed to prepare the samples for quantification via Qubit and quantitative PCR (qPCR). Upon successful qualification of the library, sequencing was conducted utilizing the Illumina NovaSeq 6000 platform (Illumina, USA). Post-sequencing, the raw sample sequences were demultiplexed based on their barcode sequences and PCR amplification sequences. The sample reads were combined using FLASH (version 1.2.11), resulting in the creation of Raw Tags. The Raw Tags were processed for quality control using the fastp software (version 0.23.4), resulting in high-quality Clean Tags. Subsequently, these Clean Tags were then compared to a database using Vsearch software (version 2.23.0) to identify and eliminate chimeric sequences, yielding the final effective data, referred to as Effective Tags. The DADA2 modules within the QIIME2 software (version 2024.2) were then utilized for noise reduction and to filter out sequences with an abundance of less than five, resulting in the generation of the Amplicon Sequence Variant (ASV) and feature datasets. Subsequently, ASV alignment was performed using the classy-sklearn module within the QIIME2 software package and the following databases: the Silva 138.1 database for the 16S rRNA gene region and the UNITE v8.2 database for the ITS region. The resulting annotations were compared to acquire species-level information for each ASV.

2.5. Statistical Analysis

To increase the statistical power of the current study, which was carried out in stands containing a single tree species, one-way analysis of variance (ANOVA) and a relatively robust post hoc multiple comparison test (Tukey’s honestly significant difference (HSD)) were applied to assess the changes in the soil physiochemical characteristics and enzymes, as well as the functionality of the bacterial and fungal communities. Differences in microbial alpha diversity among the stands with the different tree species were assessed using the Kruskal–Wallis test. The relationships among the soil physicochemical properties, soil enzyme activities, and microbial communities were examined using Spearman correlation analysis. SPSS version 27.0 (IBM Corp., New York, NY, USA) was used for all statistical analyses. The calculation of alpha diversity indices for the bacterial and fungal communities, including the Chao1, Shannon, and Simpson indices, was performed utilizing QIIME2 modules. The principal coordinate analysis (PCoA) based on UniFrac distances was conducted using R software (version 4.4.0) with the “vegan” package (version 2.6-8).

To study the relationship between the microbial community and soil environmental characteristics, a Mantel test was conducted using the “linkET” package (version 0.0.7.4) in R software to analyze the microbial species composition and soil parameters. Linear discriminant analysis (LDA) effect size (LEfSe) was employed to identify potential indicator microbiome species exhibiting significant differences between the various tree species stands and an LDA score greater than 4. The Kruskal−Wallis test was utilized to assess the significance of the differences between treatments. Furthermore, using PICRUSt2 (version 2.5.3), the potential KEGG ortholog functional profiles of the bacterial communities were determined, while the FUNGuild tool (version 1.1) was employed to categorize the nutritional types of fungi.

3. Results

3.1. Effects of Tree Species on Soil Properties

The soil physicochemical properties and enzyme activities were markedly affected by the species of trees used in afforestation during the initial phase of land-use transition from farmland to forest. Regarding the soil physical properties (

Table 2. Soil physicochemical properties in the four stands with different tree species five years after establishment.

Tree Species	Field Moisture Capacity (%)	Total Porosity (%)	Soil Bulk Density (g cm−3)	pH	SOC Content (g kg−1)	TN Content (g kg−1)	TP Content (g kg−1)	TK Content (g kg−1)	C:N	C:P	N:P
AL	278.9 ± 9.80	43.80 ± 0.55 ab	1.50 ± 0.02	6.90 ± 0.17 b	15.56 ± 0.34 ab	1.86 ± 0.06	0.55 ± 0.02	12.22 ± 0.29	8.39 ± 0.10 b	28.21 ± 0.79 b	3.36 ± 0.06 a
PA	263.88 ± 7.12	44.56 ± 1.00 a	1.51 ± 0.02	8.07 ± 0.05 a	19.01 ± 1.74 a	1.77 ± 0.22	0.52 ± 0.04	13.10 ± 0.21	11.03 ± 1.21 a	36.81 ± 2.44 a	3.39 ± 0.20 a
PD	274.93 ± 9.56	44.98 ± 0.31 a	1.47 ± 0.04	7.15 ± 0.10 b	8.99 ± 1.98 c	1.21 ± 0.21	0.49 ± 0.05	13.31 ± 0.67	7.26 ± 0.40 b	17.77 ± 2.82 c	2.41 ± 0.27 b
ZB	258.18 ± 6.72	40.89 ± 1.54 b	1.46 ± 0.02	7.16 ± 0.14 b	11.75 ± 1.38 bc	1.42 ± 0.16	0.41 ± 0.02	12.43 ± 0.20	8.25 ± 0.05 b	28.78 ± 2.49 b	3.49 ± 0.29 a

SOC: soil organic carbon; TN: total nitrogen; TP: total potassium; TK: total potassium. Values represent the mean ± SE (n = 4). Different letters within the same column indicate a statistically significant difference (p < 0.05), as determined by Tukey’s Honestly Significant Difference (HSD) test. Variables without letters showed no significant difference (p > 0.05).

), neither field moisture capacity nor soil bulk density exhibited significant changes due to afforestation. However, the total porosity was significantly greater in the bamboo PA and fast-growing tree PD stands compared to the economic tree ZB stand (p < 0.05), while no significant differences were detected between the afforestation stands and the control AL (p > 0.05). Regarding soil chemical properties, afforestation led to an increase in soil pH, with a significant enhancement observed in the PA stand compared to the control AL (p < 0.05). The soil organic carbon (SOC) levels in the PA stand were significantly greater than those in the PD and ZB stands. However, the total nitrogen (TN), total phosphorus (TP), and total potassium (TK) contents of the three forests were not significantly different compared to the control AL (p > 0.05), and no significant differences were detected among the stands with the different tree species. The C:N and C:P ratios were significantly elevated in the PA stand compared to the control AL and the PD and ZB stands (p < 0.05), whereas the N:P ratio in the PD stand was significantly lower than in the other stands (p < 0.05). In terms of soil enzyme activities (

Table 3. Soil enzyme activities in the four stands with different tree species five years after establishment.

Tree Species	Acid Phosphatase (μg g−1 2 h−1)	Sucrase (mg g−1 24 h−1)	Amylase (mg g−1 24 h−1)	Cellulase (mg g−1 72 h−1)	Urease (mg g−1 24 h−1)
AL	12.22 ± 0.29	2.65 ± 0.63	0.97 ± 0.11 a	0.29 ± 0.03 a	0.26 ± 0.01 b
PA	13.1 ± 0.21	3.64 ± 0.92	0.65 ± 0.1 b	0.35 ± 0.03 a	0.45 ± 0.10 a
PD	13.31 ± 0.67	3.51 ± 1.14	0.54 ± 0.09 b	0.20 ± 0.03 b	0.22 ± 0.03 b
ZB	12.43 ± 0.20	4.32 ± 1.53	0.82 ± 0.06 ab	0.17 ± 0.03 b	0.26 ± 0.03 b

Values represent the mean ± SE (n = 4). Different letters within the same column indicate a statistically significant difference (p < 0.05), as determined by Tukey’s Honestly Significant Difference (HSD) test. Variables without letters showed no significant difference (p > 0.05).

), no significant effects of afforestation were observed in terms of acid phosphatase and sucrase activities (p > 0.05). However, amylase activity was significantly reduced in the PA and PD stands compared to the control AL (p < 0.05). Similarly, cellulase activity was significantly lower in the PD and ZB stands compared to the AL (p < 0.05). Conversely, urease activity in the PA stand was significantly elevated compared to both the control AL and the stands with the other two tree species (PD and ZB) (p < 0.05).

3.2. Effects of Afforestation Tree Species on Microbial Community Composition

Upon examining the bacterial community composition at the phylum level, it was evident that Proteobacteria (26.53%), Acidobacteriota (20.44%), and Actinobacteriota (10.31%) demonstrated a significant predominance, each exhibiting an average relative abundance exceeding 10% (Figure 2A). Furthermore, seven additional phyla, namely Myxococcota (7.18%), Gemmatimonadota (6.05%), Chloroflexi (5.36%), Methylomirabilota (4.85%), Bacteroidota (3.89%), Latescibacterota (3.32%), and NB1.j (2.99%), displayed an average relative abundance surpassing 2%. The relative abundance of Proteobacteria and Acidobacteriota was greater in the PA stand compared to the other stands, whereas Actinobacteriota, Gemmatimonadota, and Chloroflexi exhibited a lower relative abundance in the PA stand compared to the other stands.

In fungal communities, the phyla Basidiomycota (48.56%), Ascomycota (41.80%), and Mortierellomycota (8.51%) were the only groups with a relative abundance exceeding 1%, cumulatively accounting for an average of 98.87% of the total fungal abundance (Figure 2B). The abundance of Basidiomycota was greater in the PL stand compared to the AL, whereas it was lower in the ZB stand compared to the AL. Ascomycota exhibited a greater abundance in both the PA and ZB stands compared to the AL. Conversely, Mortierellomycota showed a lower abundance in both the PA and PD stands compared to the AL.

3.3. Effects of Afforestation Tree Species on Microbial Community Diversity

In the assessment of alpha diversity within bacterial communities, the Chao1, Shannon, and Simpson indices (Figure 3A–C) were not significantly affected by the tree species employed in the afforestation during the initial transition from farmland to forest. Specifically, the Chao1 index values for the stands with the different tree species were similar (p > 0.05). In contrast, the Shannon index for the PD stand was significantly lower compared to the AL and the ZB stand (p < 0.05). Similarly, the Simpson index for the PD stand was significantly lower than that for the ZB stand (p < 0.05).

The findings from the principal coordinate analysis (PCoA) revealed a significant impact of tree species on the beta diversity of the microbial communities, with an obvious separation of the PA stand from the AL for both bacteria and fungi (Figure 4). Furthermore, Adonis analyses utilizing UniFrac distance corroborated the presence of substantial differences in the composition of the soil bacterial (adonis R² = 0.48, p < 0.001) and fungal (adonis R² = 0.26, p < 0.001) communities across the different stands, with more pronounced variations observed in the soil bacterial communities compared to the soil fungal communities.

3.4. The Association Between Soil Properties and Microbial Composition

The Mantel test results indicated that the soil organic carbon (SOC) content, total potassium (TK) content, and soil cellulase activity exhibited significant associations with the composition of the bacterial community (Figure 5). In contrast, none of the examined environmental parameters demonstrated a correlation with the composition of the fungal community. The Spearman correlation analysis revealed a statistically significant positive correlation between the SOC, TN, and TP contents and cellulase activity (Figure 6). In contrast, soil amylase activity demonstrated a statistically marked negative relationship with both the TK content and acid phosphatase activity. Furthermore, cellulase activity exhibited a positive response to variations in the soil C:N and C:P ratios.

To identify and screen indicator soil microbiome microorganisms, a Spearman correlation analysis, followed by an LEfSe analysis, was performed to explore the environmental parameters influencing these indicator microbiome microorganisms (Figure 6). The analysis revealed three predominant genera within the bacterial community: Steroidobacter and Roseisolibacter in the ZB stand, and Phenylobacterium in the PD stand. In terms of fungi, seven dominant taxa were identified: the species Chloridium submersum and Nigrospora oryzae in the AL; the genus Serendipita in the ZB stand; and within the PA stand, the order Capnodiales, the family Capnodiaceae, the genus Chaetocapnodium, and the species Chaetocapnodium philippinense were detected. The correlation analysis results for environmental parameters indicated that the SOC and TN contents, C:N and C:P ratios, and cellulase activity had a significant negative impact on the genera Steroidobacter, Roseisolibacter, and Serendipita. Conversely, the pH, C:N and C:P ratios, and cellulase activity exhibited a significant positive influence on the fungal order Capnodiales, family Capnodiaceae, genus Chaetocapnodium, and species Chaetocapnodium philippinense in the PA stand. Furthermore, the SOC, TN, and total phosphorus (TP) contents, C:N and C:P ratios, and cellulase activity were found to significantly and positively affect the fungal species Chloridium submersum in the AL.

3.5. Functional Profiles of Microbes in Stands Afforested with Different Tree Species

A PICRUSt2 analysis was used to study the soil bacterial community changes in the four stands (Figure 7A), which revealed that the Metabolism pathways KEGG annotation (61.84%–63.00%) had the highest relative abundance, followed by Genetic Information Processing (11.39%–12.87%), Environmental Information Processing (7.65%–8.45%), Cellular Processes (7.15%–7.71%), Human Diseases (6.11%–6.87%), and Organizational Systems (3.20%–3.42%). Only the abundance of the Genetic Information Processing pathway significantly differed between the PA and ZB stands (p < 0.05).

Utilizing the FUNGuild database to annotate the nutritional types of fungi in the different stands (Figure 7B), the results revealed that the predominant fungal classification in these four stands was Saprotroph. Notably, the relative abundance of Saprotrophs in the PA stand was significantly greater than in the control AL (p < 0.05). Conversely, the relative abundance of Pathotroph–Saprotroph–Symbiotroph fungi in the PA stand was significantly lower than in the AL; similarly, the ZB stands exhibited a significantly greater abundance of Pathotroph–Saprotroph–Symbiotroph fungi compared to the AL. Furthermore, the relative abundance of Pathotrophs in the ZB stand was significantly greater than in the PD stand.

4. Discussion

In this study, afforestation exerted a significant impact on both the chemical and enzymatic properties of the soil, while the physical properties remained unaffected. Notably, the effects varied among the stands with different tree species during the initial five-year growth period. Our findings indicate that afforestation with Pleioblastus amarus, Populus deltoides, or Zanthoxylum bungeanum did not result in significant alterations in the field moisture capacity, total porosity, or soil bulk density compared to the control abandoned land (Table 1). These results are consistent with those of previous research investigating the effects of land-use changes and management practices on soil characteristics. For example, the conversion of farmland into a poplar plantation over a growth period of 17 to 24 years did not significantly alter the bulk density or soil moisture [11]. The stability of the soil physical properties over this relatively short growth period indicates that the transformation of these properties following afforestation is a complex process influenced by various environmental factors and temporal dynamics [29,30,31]. Nevertheless, our findings indicate that afforestation significantly affects soil chemical properties, with variations observed among the stands with different tree species. Notably, afforestation led to an increase in the soil pH, particularly in the case of the P. amarus stand. This observation aligns with numerous studies indicating that the impact of afforestation is not uniform and can vary considerably depending on the tree species involved [11,32,33]. This variability may largely stem from the fact that different tree species contribute varying amounts of organic matter and can modify soil nutrient dynamics, consequently influencing soil pH levels [34,35]. Consistent with this, our findings demonstrated that the soil organic carbon (SOC) content was significantly affected by the tree species. Additionally, we observed significantly higher C:N and C:P ratios in the P. amarus stand compared to the abandoned land. Afforestation has been documented to result in increased SOC contents in various studies. However, the SOC content in the P. deltoides stand was significantly lower compared to that in the abandoned land. This phenomenon can be attributed to the species’ fast-growing nature. While rapid growth rates typically result in increased biomass production, they do not necessarily enhance SOC accumulation. For example, research has shown that despite substantial increases in forest floor biomass, the accumulation of organic matter in the topsoil can remain low due to slow decomposition processes [36]. Our findings demonstrate that soil enzyme activities are significantly affected by afforestation, with variations contingent upon the tree species involved. This observation aligns with prior research that has documented distinct impacts of tree species [37], and even tree genotypes [38], on soil enzyme activities. Specifically, our data indicate a significant reduction in amylase and cellulase levels in the P. deltoides stand compared to the abandoned land, while urease activity increased in the P. amarus stand. The observed decrease in amylase and cellulase activities in the P. deltoides stand is consistent with the significantly lower SOC levels associated with this species. The results highlight the critical importance of selecting suitable tree species for afforestation initiatives, given that their distinct characteristics can result in diverse impacts on soil enzyme activities and, consequently, on soil health and ecosystem functioning [39]. Collectively, these findings indicate that although afforestation can affect soil chemistry and enzymatic activity, its influence on soil physical properties may not be as significant in the short term.

The current study has shown that afforestation influences bacterial composition without affecting its alpha diversity. This result is consistent with prior research suggesting that environmental changes, such as land-use changes, can significantly alter the microbial community structure while leaving diversity metrics largely unaffected [40]. For example, a study examining the impact of green alder (Alnus viridis) encroachment demonstrated that there were considerable changes in the soil microbial communities in response to the encroachment, which were associated with specific soil properties, yet the overall diversity remained stable [41]. Conversely, our results showed that both the alpha diversity and community structure of fungi were significantly influenced by the afforestation. The community structure of the soil fungi, which play a crucial role in nutrient cycling and plant health, was shown to be influenced by the aboveground plants and environmental conditions, emphasizing the importance of considering community interactions in afforestation efforts [42]. The observed variation, wherein the bacterial alpha diversity remained constant while fungal alpha diversity was altered by afforestation, can be explained by the following factor. The introduction of novel tree species during afforestation modifies the composition of organic inputs to the soil, potentially favoring specific fungal taxa over others and resulting in alterations in fungal diversity [43]. Conversely, bacterial communities may demonstrate resilience due to their capacity to utilize a broader spectrum of substrates and their typically greater turnover rates, enabling them to sustain diversity despite alterations in environmental conditions [44]. In conclusion, the findings indicate that although bacterial communities possess the capacity to adapt to novel environmental conditions, fungal communities are likely to undergo more significant changes in diversity. This is attributable to their specialized ecological niches and interactions with the soil properties that are affected by afforestation [45]. The current study demonstrated that a greater number of fungal taxa, compared to bacterial taxa, exhibited correlations with the tree species and soil properties (Figure 6).

The bacterial community composition was markedly influenced by the soil properties, notably the SOC content and cellulase activity. It is well established in the literature that soil physicochemical characteristics, such as the SOC and nutrient levels, are pivotal in determining microbial community structures [46]. In the context of P. amarus in the current study, the elevated SOC content and cellulase activity are likely critical in providing carbon sources that influence the bacterial community structure. This aligns with the conclusion that the interaction between the soil organic carbon (SOC) content and microbial communities is essential for understanding the carbon cycling in diverse ecosystems. For example, research examining the impact of long-term fertilization on soil microbial biomass and community structure underscored the significant role of organic amendments in enhancing the SOC content and microbial activity, consequently influencing the composition of the soil bacteria [47]. Furthermore, studies on cellulose-responsive bacterial and fungal communities suggest that the presence of cellulose can result in significant alterations in the microbial community composition in diverse soil types [48]. In conclusion, elevated SOC levels and cellulase activity are crucial factors affecting the carbon supply, which influences the structure and function of bacterial communities within various ecosystems.

Regarding fungi, although no measured soil parameters were found to significantly influence their composition, distinct fungal indicator species associated with specific tree species were identified. Notably, Nigrospora oryzae, a species known to cause brown streaks, leaf spots, and latent infections in rice (Oryza sativa) [49], as well as other crops such as corn (Zea mays), wheat (Triticum aestivum), and cotton (Gossypium hirsutum) [50], was identified as one of the indicator species for the abandoned land. This indicates that the afforestation of agricultural land with trees may reduce the prevalence of specific plant pathogens associated with the previously cultivated crops. This observation aligns with prior research demonstrating that afforestation can improve soil health and potentially suppress pathogen populations [11]. Furthermore, the indicator genus for the P. amarus stand, Chaetocapnodium, which belongs to the order Capnodiales and family Capnodiaceae, is characterized by its saprobic nature and occurs as sooty molds, forming dark, superficial thalli on plant surfaces. [51]. As saprobes, the genus Chaetocapnodium could play a crucial role in the decomposition of organic matter. This role is evidenced by research indicating that interactions between saprotrophic fungi and other microbial communities can enhance decomposition rates, as these fungi are capable of utilizing root exudates and various organic compounds released into the soil [52]. Consequently, saprotrophic fungi significantly contribute to the decomposition process, thereby facilitating nutrient cycling and the turnover of organic matter. This contribution is further corroborated by the elevated SOC levels, C:N ratios, C:P ratios, and increased cellulase activity observed in the present study.

The functional predictions indicated that metabolism constitutes the principal metabolic pathway in the bacterial communities, with minimal variations between the stands with the different afforested tree species and abandoned land. This observation may be attributed to the functional redundancy and compensatory mechanisms inherent in soil bacterial populations [53,54]. According to the theory of microbial functional redundancy, while tree afforestation influenced the composition and structure of the bacterial communities, the similarities in bacterial roles and the intricate integration of ecological functions may partially obscure the functional shifts propelled by bacterial community diversity [55]. This was particularly evident during the short-term transition of land use from farmland to forest, resulting in minimal functional changes that were attributable to the afforestation.

This study investigated the changes in fungal nutritional types and determined that afforestation led to a decrease in the abundance of Saprotrophic fungi, while simultaneously boosting the prevalence of Pathotroph–Saprotroph–Symbiotroph fungi, which was particularly notable in the P. amarus stand. Saprotrophic fungi play a crucial role in the decomposition of organic matter, carbon cycling, nutrient mobilization, and soil structure formation, while in contrast, symbiotrophic fungi significantly enhance the surface area of plant roots, thereby facilitating greater access to nutrients and water for the plants in exchange for carbon [56,57,58]. The change in nutritional type from Saprotroph to Pathotroph–Saprotroph–Symbiotroph implies that afforestation with P. amarus likely transformed the farmland from non-productive yet nutrient-rich land into a productive and nutrient-rich ecosystem, thereby enhancing land productivity. This hypothesis is further supported by the relatively high urease and cellulase activities observed in the P. amarus stand soil, which indicates elevated soil carbon and nitrogen cycling [59,60]; however, no significant alterations in the SOC content or total nitrogen content were observed when compared to the abandoned land.

5. Conclusions

Following a five-year period of afforestation using bamboo (Pleioblastus amarus), fast-growing timber (Populus deltoides), or economic trees (Zanthoxylum bungeanum), our study revealed that (i) notable changes were observed in the soil chemistry and enzymatic activity compared to the control abandoned land, although the effects on the soil physical properties may not be as significant in the short term. (ii) Afforestation influenced the composition of the bacterial community, although it did not affect its alpha diversity. (iii) The increase in the SOC content and cellulase activity likely plays a crucial role in providing carbon sources that shape the bacterial community. (iv) The afforestation of farmland with trees has the potential to reduce the prevalence of certain plant pathogens, such as Nigrospora oryzae, which originated from the previously cultivated crops. (v) While afforestation did not modify the metabolic pathways of the bacterial community, it did induce a shift in the fungal community’s nutritional classification from Saprotroph to Pathotroph–Saprotroph–Symbiotroph. This transition in nutritional type implies that afforestation with P. amarus likely transformed the farmland from a non-productive yet nutrient-rich state to a productive and nutrient-rich ecosystem. As our study was based on a short growth period for these trees, how these trees influence the soil available nutrients and microbial community over a long period needs further exploration in future studies. Given that our study was conducted over a relatively short growth period of five years for these trees, and each tree species was planted in a single stand within a specific region, further research is necessary to explore the long-term effects of these trees on the soil nutrient availability and microbial communities in different stands and regions.