Alfalfa cropping is superior to cotton and rapeseed cropping in improving the quality and microbial diversity of reclaimed saline soils

ABSTRACT

Soil salinisation has led to increasing abandonment of farmlands in the arid region of northwest China, and the cultivation of cash crops has become an important way to reuse abandoned farmlands and remediate saline soils. In this study, the effects of cotton (Cot), alfalfa (Alf), and rapeseed (Rap) cropping on the bacterial and fungal communities in saline soil were compared based on high-throughput sequencing. The results showed that soil pH and electrical conductivity (EC) of the Cot, Alf, and Rap groups decreased significantly compared with those of bare land (CK). Especially, soil pH (8.3) and EC (0.93 mS cm^-1) of the Alf group decreased the most, and the contents of total nitrogen (TN) and total phosphorus (TP) increased the most, which increased by 250% and 91.4%, respectively compared with those of the CK (p < 0.05). Actinobacteriota and Proteobacteria were the dominant bacterial phyla in the three groups, Ascomycota and Basidiomycota were the dominant fungal phyla. The diversity of soil bacterial and fungal communities in the Alf group was higher than that in the Cot and Rap groups. The results of RDA and PERMANOVA analysis showed that soil pH, EC, C/N ratio, and TN were the main factors affecting soil bacterial and fungal communities in the three groups. In general, Cot, Alf, and Rap cropping all could reduce soil pH and EC and improve soil microbial community structure and diversity, but Alf cropping had the best effects in improving saline soil quality and bacterial and fungal diversity.

KEYWORDS:

• Alfalfa
• ecological restoration
• bacteria
• fungi
• saline soil

Introduction

Xinjiang is located in the arid region of China. The total area of oases accounts for 6% of Xinjiang’s total area, but over 95% of Xinjiang’s population lives in the oases. Therefore, the soil quality in the oases is of great significance for local food security. However, due to the arid climate and improper irrigation methods, a large number of farmlands in the oases have been salinised (Cheng et al. 2018a). Soil salinisation and secondary salinisation have become the main factors leading to soil degradation and limiting the increase of agricultural production (Hassani et al. 2020). At present, vegetation restoration has been regarded as an effective means for the remediation of salinised soils (Wang et al. 2016). It not only improves the physicochemical properties of salinised soils, but also promotes the stability of ecosystems.

Soil physicochemical properties are the main focus of soil degradation and restoration researches (Yang et al. 2016; Zhang et al. 2017; Cheng et al. 2018b). Generally, different agronomic measures can lead to changes in soil physical and chemical properties (Marshall et al. 2011). Besides, by changing the composition and diversity of crop root exudates, crop residues, and symbiotic links, soil carbon input, nutrient availability, and soil structure can also be directly changed (Ai et al. 2018). With the deepening of research in soil microecology, the roles of soil microorganisms in vegetation restoration and soil quality evaluation have been widely studied by scholars (Rashid et al. 2016). For example, Xu et al. (2021) found that different vegetation types and cropping regimes have different effects on soil physicochemical properties, which further leads to differences in soil microbial diversity and composition and soil ecosystem functions. Xinjiang is the most important cotton production base of China, and many abandoned salinised farmlands in Xinjiang have been reclaimed to plant cotton in recent years. Therefore, the previous research on the reclamation of abandoned salinised farmlands mainly focuses on cotton planting. For example, Cheng et al. (2018a) reported that soil bacterial community structure and diversity showed significant difference under different cropping regimes (maise-wheat rotation, tree-cotton intercropping, and monoculture cotton) during reclamation of abandoned salinised farmland, but Firmicutes was the dominant phylum under the three cropping regimes. Yang et al. (2018) reported that the duration of continuous cropping obviously influenced soil bacterial community composition and structure in cotton fields, and the first 10 years of cotton cropping significantly increased soil bacterial diversity. Zhang and Du (2018) reported that the abundance of Actinobacteria, Acidobacteria, Firmicutes, and Nitrospirae were obviously changed by continuous cotton cropping in reclaimed saline soils. However, cotton cropping cannot continuously improve the quality of salinised soil, and long-term continuous cotton cropping may negatively affect soil microbial communities and even lead to a decline in soil quality. For example, Ma et al. (2023) reported that long-term (≥8 years) continuous cropping interfered with the complexity of soil bacterial co-occurrence networks and reduced collaboration between OTUs. Besides, Xi et al. (2019) reported that long-term continuous cropping increased the incidence of Verticillium wilt and led to a decline in cotton yield and quality.

Therefore, there is an urgent need to select the optimal crop that can meet the needs of Xinjiang’s agricultural industry and the remediation of Xinjiang’s salinised soil. As a leguminous plant, alfalfa can fix soil nitrogen, improve soil quality, and increase crop yield; Besides, it can also be used as forage to alleviate the forage shortage in Xinjiang. Although there have been some studies on improving the quality of saline soil by alfalfa cropping, they mainly focus on soil chemical properties and bacterial communities (Elgharably and Benes 2021; He et al. 2022). There is still a lack of research on soil fungal communities during the reclamation of abandoned salinised soil by alfalfa cropping. Therefore, in this study, three salinised soil reclamation patterns (cotton (Cot), alfalfa (Alf), and rapeseed (Rap) cropping) were selected, and their effects on soil bacterial and fungal communities in salinised abandoned farmlands were explored and compared based on high-throughput sequencing. The purposes were to explore (1) Whether the effects of alfalfa cropping on soil physicochemical properties as well as bacterial and fungal communities were better than those of cotton and rapeseed cropping in the reclamation of abandoned salinised farmland; and (2) the possible factors driving soil bacterial and fungal community changes under alfalfa cropping. This study will provide reference for the remediation and utilisation of salinised soils in arid areas.

Materials and methods

Study site

The study area is located in Daquangou Town, Shihezi City (44°37ʹ N, 86°08ʹ E, 451 m a.s.l.) in northwest China. This region has a temperate continental climate, with long winters and short summers. The average annual precipitation was 125∼208 mm, the average annual evaporation was 2005mm, and the frost-free period was 168∼171 days. The soil type was grey desert soil. Due to the long-term use of saline water for irrigation, the high groundwater table, and the strong evaporation, soil salt increasingly accumulate in the surface soil, leading to widespread soil salinisation (secondary salinisation) and the abandonment of many farmlands. The dominant species were Tamarix chinensis Lour., Kalidium foliatum Moq., and Salsolacollina Pall.

Experimental design

Abandoned salinised farmlands were selected to carry out vegetation restoration since 2018. There were four groups in this experiment, including (1) CK, natural restoration; (2) Cot group, cotton (Gossypium hirsutum L) was cropped; (3) Alf group, alfalfa (Medicago L.) was cropped; Rapeseed (Brassica napus L.) was cropped. The area of each group was 1 ha. Nitrogen (300 kg·ha), phosphorus (120 kg·ha), and potassium (60 kg·ha) fertilisers were basally applied through the drip application system after dissolving in water. Each group was irrigated for 8 ∼ 10 times in the growing season. The total irrigation amount was 4500 m³·ha.

Sampling

In mid-July 2021, 50 soil samples (0∼20 cm in depth, 5 cm in diameter) were randomly collected in each group using soil drills, and fully mixed. The soil samples were placed in ice boxes and transported to the laboratory. After removing stones and plant residues, soil samples were sieved through a 2 mm sieve. Finally, the soil samples were divided into two parts. One part was used for the determination of soil physicochemical properties, and the other part was stored at −80°C for high throughput sequencing (three per group).

Determination of soil physicochemical properties

Soil pH was measured with a composite electrode (INESA Scientific PHSJ-3F), and the ratio of soil to water was 1: 2.5. Soil EC was measured with a conductivity meter (Orion Versa Star Pro, VSTAR20) using a soil/water suspension (1: 5). Soil total phosphorus (TP) content was determined using a spectrophotometer (UV5800, Shanghai Metash Instruments, China) based on the HClO₄-H₂SO₄ method, soil total potassium (TK) content was determined using a flame photometer (F-500, Shanghai Metash Instruments, China) based on the NaOH melting method, and soil total nitrogen (TN) and total carbon (TC) contents were determined using an automatic element analyzer (German element analyzer, Vario MACRO cube) (Collins et al. 2019).

Metagenomic sequencing and gene analysis

Soil bacterial diversity was determined according to the method of Wang et al. (2019). Total DNA (0.25 g) was extracted from soil samples using the FastDNA^® Spin Kit for Soil according to the protocol. DNA purity, concentration, and integrity were detected using NanoDrop2000, picogreen, and agarose gel electrophoresis, respectively. The hypervariable region of bacterial 16S rRNA gene V4 (GTGCCAGCMGCGG) – V5 (CCGTCAATTCMTTTRAGTTT) was amplified with 515F/907R primer. PCR reaction solution (20 μL) was composed of 4 μL of 5 × FastPfu Buffer, 2 μL of 2.5 mM dNTPs, 0.8 μL of Primer (5 μM), 0.4 μL of FastPfu Polymerase, and 10 ng of template DNA. The PCR products were detected using a 2% agarose gel, followed by purification and quantification with the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA) and QuantiFluor-ST™ (Promega, USA), respectively.

Soil fungal diversity was determined according to the method of Wang et al. (2019). Total DNA was extracted from 0.25 g sample using the Power Soil DNA Isolation kit (MOBIO Laboratories, Inc., USA), and then the fungal ITS1 region was amplified using primers ITS1F (5ʹ-CTTGGTCATTTAGAGGAAGTAA-3ʹ) and ITS2R (5ʹ-GCTGCGTTCTCTCTCCATCGATGC-3ʹ). The 20 μL reaction system for amplification was composed of 2 μL of 10 × Buffer, 2 μL of 2.5 mmoL/L dNTPs, 0.8 μL of ITS1F (5 μmoL/L), 0.8 μL of ITS2R (5 μmoL/L), 0.2 μL of TaKaRa rTaq DNA Polymerase, 0.2 μL of BSA, 10 ng of template DNA, and ddH₂O. Each sample had three replicates. The PCR products of the same sample were mixed and detected using 2% agarose gel electrophoresis. The AxyPrep DNA gel recovery kit (AXYGEN Company) was used to collect the PCR products, and the products were subjected to quantitative detection using the QuantiFluor for PCR products™ – ST Blue Fluorescence Quantitative System (Promega Company). Finally, the Miseq sequencing library was constructed.

The high-throughput sequencing of bacteria (V4-V5) and fungi (ITS1) samples were conducted by Shanghai Meiji Biomedical Technology Co., Ltd. using the Illumina MiSeq PE300 (Illumina, Inc., San Diego, CA, USA). Sequences were filtered through QIIME (version 1.9.0), and then the chimeric sequences were removed using the Mothur software to obtain high-quality sequences for subsequent analysis. The operational taxon unit (OTU) was clustered by UPARSE with a similarity of 97%, and the chimeric sequences were identified and deleted by UCHIME. The ribosomal database project (RDP) classification of each 16S rRNA gene sequence was performed using 70% confidence threshold against the Silva (SSU123) 16S rRNA database.

Statistical analysis

Venn map was drawn to compare the similarity between OTU-level soil bacterial and fungal communities. According to the species abundance of each sample in the OTU list, Mothur software was used to calculate the diversity index (Shannon) and richness index (Chao) of soil bacteria and fungi. Principal component analysis (PCA) was used to reveal the changes of soil bacterial and fungal community structure. Variance inflation factor (VIF) analysis was used to select autocorrelated soil properties. The effects of soil properties on soil bacterial and fungal communities were examined by Mantel test, redundancy analysis (RDA), and canonical correspondence analysis (CCA). Permutation multivariate analysis of variance (PERMANOVA) was used to decompose the total variance by Bray Curtis distance, to analyse the explanatory degree of different soil properties on the sample difference, and the Permutation test was performed to test significance. The Spearman correlation coefficient was calculated to determine the correlation between soil properties and soil bacterial and fungal communities. R software was used to perform PCA, ANOSIM, VIF, PERMANOVA, RDA, and correlation analyses.

Results

Changes in soil properties

The soil pH and EC of the Cot, Alf, and Rap groups decreased significantly compared with those of the CK group (p < 0.05) (Table 1), and the Alf group held the lowest soil pH (8.3) and EC (0.93 mS cm⁻¹). The TP and TN contents of the Alf group were the highest, increasing by 91.4% and 250%, respectively compared with those of the CK group (p < 0.05); Besides, the TK content of the Alf group increased by 48.6% (p < 0.05). The soil C/N ratio of the Cot, Alf, and Rap groups decreased by 3.1%, 5.8%, and 4.7%, respectively compared with that of the CK group (p < 0.05).

Table 1. Soil properties under the reclamation of abandoned salinised farmlands by planting different crops.

Samples	pH	EC (mS·cm^−1)	TK (g·kg^−1)	TP (g·kg^−1)	TN (g·kg^−1)	TC (g·kg^−1)	C/N ratio
CK	9.13 ± 0.08a	1.64 ± 0.13a	57.64 ± 2.81c	0.81 ± 0.05d	0.18 ± 0.01d	2.05 ± 0.12c	11.39 ± 1.52a
Cot	8.37 ± 0.05b	1.12 ± 0.06b	80.89 ± 2.27b	0.92 ± 0.05c	0.28 ± 0.01c	2.20 ± 0.16b	7.86 ± 0.61b
Alf	8.30 ± 0.14b	0.93 ± 0.038c	85.68 ± 4.18b	1.55 ± 0.05a	0.63 ± 0.02a	3.04 ± 0.11a	4.83 ± 0.25b
Ra	8.48 ± 0.08b	1.00 ± 0.03bc	114.46 ± 6.11a	1.25 ± 0.05b	0.47 ± 0.01b	2.82 ± 0.16b	6.00 ± 0.23c

Notes: Cot, Cotton cropping; Alf, alfalfa cropping; Rap, rapeseed cropping; EC, electrical conductivity; TK, total potassium; TP, total phosphorus; TC, total carbon; TN, total nitrogen; C/N ratio, total carbon/total nitrogen ratio; CK, abandoned farmland. Different lowercase letters in each column indicate significant difference between groups (Tukey’s test, p < 0.05).

Composition and diversity of soil microbial communities

Actinobacteriota, Proteobacteria, Chloroflexi, Acidobacteriota, Gemmatimonadota, and Firmicutes were the dominant bacterial phyla (relative abundance >5%) of the Cot, Alf, and Rap groups (Figure 1(a)), and Actinobacteriota (44.44%) was the dominant bacterial phyla of the CK group. Besides, the relative abundance of Actinobacteriota of the Cot, Alf, and Rap groups decreased significantly compared with that of the CK group, among which the Alf group had the lowest relative abundance of Actinobacteriota (27.13%). Acidobacteriota and Gemmatimonadota were the dominant bacterial phyla of the Alf group, and their relative abundances were 20.82% and 6.11%, respectively. Firmicutes was the dominant bacterial phyla of the Rap group, and the relative abundance was 6.52%.

Figure 1. Relative abundance of dominant bacterial (a) and fungal phyla (b) (relative abundance >5%) under the reclamation of abandoned salinised farmlands by planting different crops. Cot, Cotton cropping; Alf, alfalfa cropping; Rap, rapeseed cropping. The same below.

Ascomycota, Basidiomycota, Mortierellomycota, and Rozellomycota were the dominant fungal phyla of the Cot, Alf, and Rap groups (Figure 1(b)), and Ascomycota was the dominant fungal phylum of the CK group. The relative abundance of Ascomycota of the Alf group was 14.76% and 15.07% higher than that of the Cot and Rap groups, respectively (p < 0.05). The relative abundance of Rozellomycota of the Alf group was higher than that in the other groups (p < 0.05), but the relative abundance of Basidiomycota of the CK group was 28.96% and 15.16% higher than that of the Alf and Rap groups, respectively (p < 0.05). Besides, the relative abundance of Mortierellomycota of the Rap group was higher than that of the Cot, Alf, and CK groups (p < 0.05).

Rubrobactor, Blastococus, and Geodermatophilus were the dominant bacterial genera of the CK group. Arthroactor, RB41, and Bacillales were the dominant bacterial genera of the Cot, Alf, and Rap groups, respectively. Besides, there were differences in the relative abundance of Gibberella, Neocamarosporium, Laburnicola, Sporormiella, Chaetomium, Neocosmopora, Gibellulopsis, Chordomyces, Mortierella, and Filoasidiales between groups (p < 0.05).

After reclamation, the soil bacterial and fungal community structure changed significantly (Figure 2). The results of cluster analysis showed that the samples of the CK group were clustered together, and those of the Cot, Alf, and Rap groups were clustered together (Figure 2(a,b)). The results of PCA analysis showed that there were significant differences in bacterial (R²= 0.835; P = 0.001) and fungal (R²= 0.793; P = 0.001) classification between groups (Figure 2(c,d)).

Figure 2. Hierarchical clustering analysis (a, bacteria; b, fungi) and PCA analysis (c, bacteria; d, fungi) of bacterial phyla and fungal phyla at the OTU level.

According to the results of OTU classification of bacterial communities (Figure 3), the total OTU number of the four groups was 1090. Unique OTUs were the most in the Alf group (584) and the least in the Rap group (179). According to the results of OTU classification of fungal communities, the total OTU number of the four groups was 71. Unique OTUs were the most in the CK group (224) and the least in the Cot group (65).

Figure 3. Venn diagram of unique and common OTUs.

The diversity indexes of bacterial communities of the Cot, Alf, and Rap groups increased significantly compared with those of the CK group. The Shannon index of the Alf group increased compared with that of the other groups (p < 0.05) (Figure 4(a)). The Chao index of the Cot, Alf, and Rap groups increased compared with that of the CK group (p < 0.05), and there was no difference between the Cot and Alf groups (p > 0.05) (Figure 4(b)). The Shannon index of fungal communities of the Cot, Alf, and Rap groups decreased compared with that of the CK group (p < 0.05), and the Shannon index of the Alf group increased compared with that of the Cot and Rap groups (p < 0.05) (Figure 4(c)). In addition, the Chao index of the Alf group increased compared with that of the other groups (p < 0.05) (Figure 4d).

Figure 4. α-diversity of soil bacterial (a, b) and fungal communities (c, d) during the reclamation of abandoned salinised farmlands by planting different crops. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Correlation between soil microbial communities and soil properties

The soil properties including TP, C/N ratio, TK, and EC were selected through VIF analysis and used for RDA analysis. The results of RDA analysis showed that EC (R²= 0.76, p = 0.007), TP (R²= 0.68, p = 0.01), and C/N ratio (R²= 0.81, p = 0.002) explained 70.35% of the variation in bacterial communities at the phylum level (Table 2, Figure 5(a)). TK, EC (R²= 0.93, p = 0.004), C/N ratio (R²= 0.91, p = 0.001), and TP (R²= 0.67, p = 0.013) explained 70.35% of the variation in fungal communities at the phylum level (Table 3, Figure 5(b)). EC and C/N ratio had a more pronounced impact on fungal communities than on bacterial communities (Tables 2, 3). The results of correlation analysis showed that at the phylum level, for bacterial communities, the relative abundances of Actinobacteriota, Acidobacteriota, Gemmatimonauta, Myxococcota, and Methylomirabilota were mainly related to the changes of soil factors (p < 0.05) (Figure 6(a)). For fungal communities, the relative abundances of Mortierellomycota, Rozellomycota, Glomeromycota, and Olpidiomycota were mainly related to the changes of soil factors (p < 0.05) (Figure 6(b)).

Figure 5. Redundancy analysis (RDA) between bacterial community and soil properties (a) as well as canonical correspondence analysis (CCA) between fungal community and soil properties (b). EC, electrical conductivity; TK, total potassium; TP, total phosphorus; C/N ratio, total carbon/total nitrogen ratio.

Figure 6. Correlation between soil factors and bacterial (a)/fungal (b) community at the phyla level. TC, total carbon; TN, total nitrogen. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Table 2. Correlation between soil factors and RDA axes.

Soil factor	RDA1	RDA2	R^2	p-value
EC	−0.8065	−0.5913	0.7656	0.007
TK	−0.9879	0.1551	0.0612	0.756
TP	−0.0659	0.9978	0.6806	0.01
C/N ratio	−0.6746	−0.7382	0.8082	0.002

Notes: EC, electrical conductivity; TK, total potassium; TP, total phosphorus; C/N ratio, total carbon/total nitrogen ratio.

Table 3. Correlation between soil factors and CCA axes.

Soil factor	CCA1	CCA2	R^2	p-value
EC	−0.9976	0.069	0.9268	0.004
TK	0.6739	0.7389	0.0113	0.929
TP	0.637	−0.7709	0.6655	0.013
C/N ratio	−0.9671	0.2546	0.916	0.001

Notes: EC, electrical conductivity; TK, total potassium; TP, total phosphorus; C/N ratio, total carbon/total nitrogen ratio.

At the genus level, the results of PERMANOVA analysis further confirmed the significant impacts of soil factors on soil bacterial and fungal communities of the Cot, Alf, and Rap groups (Table 4). Although the VIF analysis excluded pH, TN, and TC, the results of PERMANOVA analysis showed that pH, EC, C/N ratio, and TN had a significant impact on the change of bacterial and fungal communities at the genus level (p < 0.01).

Table 4. PERMANOVA analysis of the effects of soil factors on soil bacterial and fungal community.

Classification	Soil factor	Sums of Sqs	Mean Sqs	F Model	R^2	p-value
Bacteria	pH	0.766	0.766	8.413	0.457	0.001
	TN	0.610	0.610	5.712	0.364	0.001
	C/N ratio	0.718	0.718	7.494	0.428	0.001
	EC	0.703	0.703	7.213	0.419	0.004
	TP	0.484	0.484	4.057	0.289	0.008
	TC	0.250	0.250	1.752	0.149	0.171
	TK	0.205	0.205	1.391	0.122	0.252
Fungi	pH	1.266	1.266	5.709	0.363	0.001
	TN	1.064	1.064	4.396	0.305	0.001
	C/N ratio	1.208	1.208	5.306	0.347	0.001
	EC	1.257	1.257	5.645	0.361	0.002
	TP	0.893	0.893	3.447	0.256	0.011
	TK	0.843	0.843	3.191	0.242	0.013
	TC	0.870	0.870	3.329	0.250	0.013

Notes: EC, electrical conductivity; TK, total potassium; TP, total phosphorus; TC, total carbon; TN, total nitrogen; C/N ratio, total carbon/total nitrogen ratio; Sums of Sqs, Sum of squared deviations; Mean Sqs, Mean variance; F. Model, F test value; R², Interpretation of soil factors for sample differences; p-value, confidence level. Bold p values indicate p < 0.05.

Discussion

Alfalfa cropping is superior to cotton and rapeseed cropping in improving soil chemical properties

Cropping and agricultural management have great impacts on soil chemical properties, which in turn greatly affects agricultural productivity (Cheng et al. 2018b). In this study, the soil pH and EC decreased under the three remediation modes compared with those of the CK, and those of the Alf group decreased the most. The reasons are as follows: (1) the leaching of irrigation water leads to a decrease in the content of soluble salt ions in the surface soil (Yang et al. 2016). (2) Alfalfa nodulating rhizobia improves the stress resistance of alfalfa plants by increasing antioxidant enzyme activities and the content of osmolytes such as proline, betaine, and soluble proteins (Al-Farsi et al., 2020). (3) During vegetation restoration, crop root exudates, organic matter mineralisation, and organic acids secreted by soil microorganisms could reduce soil pH (Franco-Otero et al. 2012). This study also found that the soil TK, TP, TN, and TC contents of the Cot, Alf, and Rap groups increased compared with those of the CK group (p < 0.05). This is also beneficial for reducing soil pH. The increase of soil TK, TP, TN, and TC content can significantly change the nutrient uptake of crops and soil structure, thereby reducing soil pH (Ding et al. 2017). It should be noted that under the same fertilisation regime, there were significant differences in the soil TP, TN, and TC contents between Cot, Alf, and Rap groups. This may be caused by differences in the soil nutrient utilisation of different crops (Weih et al. 2018). Besides, it was also found that the soil TN content of the Alf group was higher than that of the Cot and Rap groups. This further confirms that rhizobia on Alfalfa roots could fix N (Du et al. 2019). Taken together, this study proposes that Alfalfa can be cropped in abandoned salinised farmlands to increase soil TN content, reduce soil pH and EC, and improve soil quality.

Alfalfa, cotton, and rapeseed cropping on abandoned salinised farmland significantly change soil bacterial and fungal communities

Different crops have different soil nutrient uptake and root exudates, causing differences in soil microbial community structure and diversity (Ishaq et al. 2017). However, in this study, it was found that Actinobacteria, Proteobateria, Chloroflexi, and Acidobacteriota were the common dominant species in the Cot, Alf, and Rap groups. These dominant bacteria have also been observed in the soils of other farmlands (Hua et al. 2016; Cheng et al. 2018a). Although the soil pH and EC decreased significantly after the reclamation of abandoned salinised farmland, the farmland was still mildly salinised, and the soil nutrient content was still lower than that of non-salinised farmland (Table 1), which made Actinobacteria and Proteobateria occupy an ecological niche advantage in the bacterial community. Delgado-Baquerizo et al. (2017) reported that Actinobacteria was able to decompose and utilise organic compounds that were difficult to decompose such as cellulose, lignin, and chitin in saline soil. Zhang et al. (2017) reported that Proteobateria can degrade polysaccharides in oligotrophic arid environments. In this study, the high abundance of Actinobacteria and Proteobateria was detected in the Cot, Alf, and Rap groups. This makes the niche advantage of bacteria such as Actinobaciota and Proteobateria that can adapt to the stress environment more obvious. It was worth noting that the relative abundance of Acidobacterota (20.82%) was the highest in the Alf group. This may be due to the fact that the contents of soil TC and TN of the Alf group are higher than those of the other groups. Acidobacterota has been shown to be highly related to the degradation of organic substances. Therefore, all above results indicate that Alfalfa cropping in abandoned salinised farmland is not only conducive to the enrichment of bacteria and the degradation of organic matter, but also conducive to increasing soil TC and TN content.

Ascomycota can degrade plant litters by secreting holocellulase (Yin and Zhang 2022). In this study, among soil fungi, Ascomycota had the highest relative abundance in the four groups, and the relative abundance of Ascomycota of the Alf group was 14.76% and 15.07% higher than that of the Cot and Rap groups, respectively. This may be due to the fact that the residual alfalfa stems and leaves on the surface and root activities after mowing and harvesting provide good conditions for the growth and reproduction of Ascomycota. Basidiomycota can decompose complex organic compounds, such as polycyclic aromatic hydrocarbons, and play an important role in nutrient uptake and decomposition and carbon fixation in agricultural soils (Syed et al. 2010). Besides, Basidiomycota is more adaptable to changes in the soil environment brought about by farmland disturbances than Ascomycota (Yin and Zhang 2022). This study found that the relative abundance of Basidiomycota in the Cot group was 28.96% and 15.16% higher than that in the Alf and Rap groups, respectively, and Ascomycota was the dominant fungus in the Alf group. This may be due to that Alfalfa is a perennial plant, the soil is therefore less disturbed than cotton- and rapeseed-field soils. However, straw incorporation causes frequent soil disturbance during cotton cropping, which is conducive to increasing Basidiomycota abundance.

The results of cluster analysis and PCA showed that the soil bacterial and fungal community structure of the Alf, Cot, and Rap groups changed significantly, compared with those of the CK group, and there was also a significant difference between Alf, Cot, and Rap groups. The analysis results of the number of unique OTUs in the Alf, Cot, and Rap groups also proved this difference, that is, the unique OTUs of bacterial communities were the most in the Alf group (584) and the least in the Rap group (179). The unique OTUs of fungal communities were the most in the CK group (224) and the least in the Cot group (65). This difference is due to changes in the abundance of bacterial and fungal species in the soils discussed above. Besides, the changes in soil bacterial communities were also reflected in the diversity of the communities (Figure 4).

Microbial diversity and its function in nutrient transformation are essential for sustainable agricultural systems (Zhao et al. 2014). In this study, the diversity indexes of bacterial communities of the Cot, Alf, and Rap groups increased significantly compared with those of the CK group. This may be due to that the decrease of soil pH and EC after reclamation is conducive to the growth of soil bacteria, leading to an increase in soil bacterial diversity. It is generally believed that bacterial diversity decreases with the increase of soil salinity and pH (Rath et al. 2019). This study also found that the Shannon index of the Alf group was higher than that of the other groups (Figure 4(a)). This may be due to the greater decrease in soil pH and EC in the Alf group (Table 1). In addition, fertilisation, root exudates, and straw incorporation during reclamation can improve the availability of soil nutrients, which further improves soil bacterial diversity (Su et al. 2015). Although Cot, Alf, and Rap cropping led to a decrease in Shannon index of fungal community, but the Shannon index of the Alf group was higher than that of the Cot and Rap groups (Figure 4(c)). Therefore, Alfalfa cropping is superior to cotton and rapeseed cropping in improving bacterial and fungal diversity in salinised soils.

Soil C/N ratio is an important factor affecting the bacterial and fungal communities in salinised soil

Previous studies have shown that soil physicochemical properties, especially soil pH and EC, greatly affect soil microbial community composition (Neupane et al. 2019). This study obtained similar results, i.e. soil bacterial and fungal communities were greatly affected by soil pH and EC in the Cot, Alf, and Rap groups. The relative abundance of Actinobacteria, Acidobacteriota, Gemmatimonadota, Bacteroidota, and Myxococcoda was significantly correlated with soil EC and pH (Figure 6(a)), while that of Rozellomycota and Glomeromycota was significantly negatively correlated with soil EC and pH (Figure 6(b)). Moorhead and Sinsabaugh (2006) reported that soil organic matter bioavailability and soil nutrient content can be characterised by C/N ratio. In this study, soil C/N ratio had a significant effect on soil bacterial and fungal communities, and explained 42.80% and 34.70% of the variations of bacterial and fungal communities, respectively. Besides, it was also found that soil C/N ratio was correlated with the relative abundances of dominant bacteria Actinobacteria, Acidobacteriota, and Gemmatimonadota and dominant fungi Rozellomycota and Glomeromycota. This further supports the idea that soil C/N ratio is an important driver behind the difference in soil microbial communities between groups (Chávez-Romero et al. 2016).

Conclusions

In this study, cotton, alfalfa, and rapeseed cropping were employed for the reclamation of abandoned salinised farmlands in the arid regions of northwest China. It was found that the soil pH and EC decreased the most after alfalfa cropping, which was conductive to increasing the content of soil total nitrogen and carbon. The results of high-throughput sequencing showed that after alfalfa cropping, the diversity of soil bacterial and fungal communities was higher than that after cotton and rapeseed cropping, and Aconobacteriota, Gemmatimonadota, and Ascomycota were the dominant bacterial and fungal communities (relative abundance >5%). Soil pH, EC, total nitrogen, and C/N ratio were the key factors affecting soil bacterial and fungal communities after cotton, alfalfa, and rapeseed cropping. In a word, alfalfa cropping had the most significant improvement effect on soil bacterial and fungal diversity and soil quality in abandoned salinised farmlands. Besides, as a global forage crop, alfalfa can be planted for ecological restoration of saline land in the future, which can improve soil quality, increase soil fertility, and improve the natural environment.

Acknowledgements

This research was financially supported by the High-level Talents Research Start-up Project of Shihezi University (grant number: RCZK201928).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the High-level Talents Research Start-up Project of Shihezi University [grant number RCZK201928].

Notes on contributors

Fating Yin

Fating Yin, a PhD graduate from the Agricultural College at Shihezi University, is currently a researcher with Laboratory of Resources and Environmental Carrying Capacity of the Yellow River Basin at Zaozhuang University. His research interests include remediation of saline soils, crop cultivation, and soil microorganisms.

Lei Yang

Lei Yang is a researcher with Key Laboratory of Oasis Eco-agriculture of Xinjiang Production and Construction Corps and currently a doctoral candidate at the Agricultural College of Shihezi University. His research interests include crop cultivation, straw incorporation, and soil microorganisms.

Wei Pang

Wei Pang, a PhD graduate from the Agricultural College at Shihezi University, is currently a researcher with Key Laboratory of Oasis Eco-agriculture of Xinjiang Production and Construction Corps. His research interests include remediation of saline soils, crop cultivation, and soil microorganisms.