Research on soil bacterial community assembly and function under different straw returning practices in arid and semi-arid agricultural ecosystems over multiple years

Abstract

Introduction: Straw return has gained attention for its potential to improve soil quality and crop yields, particularly in semi-arid regions like the Tumu Chuan Plain Irrigation Area. Soil bacteria play a crucial role in regulating soil biological processes, and understanding how straw return affects bacterial populations can guide better agricultural management practices.

Methods: We investigated the impact of continuous straw return on soil bacterial communities using 16S rRNA gene sequencing. Four treatments were applied: Farmers' shallow rotation (CK), straw incorporated with deep tillage (DPR), straw incorporated with subsoiling (SSR), and no-tillage mulching straw return (NTR). Bacterial community structure, metabolic pathways, and assembly mechanisms were analyzed using Bugbase and PICRUSt2 for phenotypic and metabolic pathway predictions.

Results: The study found that straw return practices significantly altered the relative abundance and life history strategies of bacterial phyla, mainly influenced by soil organic matter (SOM) and enzyme activity. The K-strategist to r-strategist ratio was highest in CK (2.06) and lowest in SSR (1.89). DPR and NTR treatments significantly changed bacterial community structure compared to CK (p < 0.05), resembling SSR. Predictions showed that DPR and NTR enhanced carbohydrate and amino acid metabolism and promoted more stable bacterial networks, with homogenous selection and drift effects. Bacterial aggregation in all treatments was driven by random processes, with varying aggregation levels: CK (20%), DPR (38.6%), SSR (16.5%), and NTR (30.7%).

Discussion: The study demonstrates that continuous straw return practices significantly impact soil bacterial communities. DPR and NTR notably improved microbial diversity, bacterial cooperation, and ecosystem stability. These findings provide valuable insights for sustainable agricultural practices in semi-arid regions, enhancing soil microbial ecology and soil health through strategic straw return.

Keywords: bacterial lifestyle; ecosystem stability; semi-arid agriculture; soil microbial community assembly; straw return practices.

Figure 1

Composition and diversity of soil bacterial communities. (A) Differences in relative abundance of soil bacterial communities at phylum and genus levels among different straw-returning practices. (B) Venn diagram of bacterial OTUs in CK, DPR, SSR, and NTR treatments. (C) Error lines for bacterial α-diversity in CK, DPR, SSR, and NTR treatments are standard deviations (SD) from the mean. Different lowercase letters indicate significant differences among the three treatments (p < 0.05). (D) Non-metric multidimensional scaled ordination plots of loam bacterial communities for CK, DPR, SSR, and NTR treatments were tested for significance using Adonis and Anosim tests (^**p < 0.01). CK, farmer’s shallow rotary; DPR, deep rotary straw return; SSR, straw incorporated with subsoiling, and NTR, no-tillage mulching straw return.

Figure 2

Soil bacterial phenotypes and life history characteristics. (A) Bacterial community prediction phenotypes obtained with Bugbase in CK, DPR, SSR, and NTR treatments. (B) G+:G− ratios in CK, DPR, SSR, and NTR soils. (C) Relative abundance of K-strategy and r-strategy bacteria in CK, DPR, SSR, and NTR soils. The error line is the standard deviation (SD) of the mean. Different lowercase letters indicate significant differences (p < 0.05) among the three treatments. G+:G− ratio: phenotypic ratio of Gram-positive to Gram-negative bacteria, K-strategy: relative abundance of representative K-strategy clades (Actinobacteriota, Chloroflexi, and Acidobacteriota), r-strategy: relative abundance of representative r-strategy clades (Proteobacteria, Firmicutes, and Gemmatimonadota) relative abundance.

Figure 3

Assembly processes and symbiotic networks of soil bacterial communities. (A) β-Nearest taxonomy index (βNTI) of CK, DPR, SSR, and NTR soil bacteria with different lower-case letters indicate significant differences among the three treatments (p < 0.05). (B) Relative contribution of deterministic and stochastic assembly processes in MM, MP, and MS soils. (C) Symbiotic network nodes of bacterial communities in MM, MP, and MS soils indicate individual operational taxonomic units (OTUs), while edges indicate significant correlations between OTUs. The colors of the nodes indicate the main gates of K-strategy and r-strategy, and gray nodes indicate other gates. The size of the nodes is proportional to the number of connections. Only nodes that are significantly (p < 0.05) and strongly correlated (Spearman correlation coefficient >0.5 or <−0.5) are connected (edges). The thickness of the edge between two nodes is proportional to the value of the Spearman correlation coefficient. Pink and blue edges indicate positive and negative interactions between two individual nodes, respectively. (D) Network statistics including average clustering coefficient, average degree, graph density and ratio of positive and negative edges. (E) ZiPi analysis to filter out key nodes. Hes, heterogeneous selection; HoS, homogeneous selection; DL, diffusion limit; HD, homogeneous dispersal; DR, drift.

Figure 4

Drivers of bacterial community composition, life history and assembly processes. (A) Relationships between bacterial communities, life history strategies and soil properties based on Pearson’s correlation and Mantel test. (B) Relative importance of soil physicochemical and microbiological properties on bacterial assembly processes in CK, DPR, SSR, and NTR treatments. The upper panel shows the variable predicted importance (VIP) values and the lower panel shows the partial least squares (PLS) standardized coefficients (±SEM). The error line is the standard deviation (SD) of the mean. Higher VIP values indicate greater importance in determining the assembly process. Blue dots indicate VIP values greater than 1. PLS standardized coefficients show the direction and magnitude of each variable’s impact on the assembly process. Variables with blue bars are significant (p < 0.05) where the VIP value is greater than 1. System status: BD, soil bulk density; SM, soil moisture content; AN, soil dissolved alkaline nitrogen; SOM, soil organic matter; AP, soil quick phosphorus; ALP, alkaline phosphatase; Glu, glutamine synthetase; H₂O₂, catalase.