Spatial Heterogeneity in Soil Microbial Communities Impacts Their Suitability as Bioindicators for Evaluating Productivity in Agricultural Practices

Abstract

Soil microorganisms are increasingly recognized as critical regulators of farmland soil fertility and crop productivity. However, the impacts of spatial heterogeneity in soil microbial communities on bioindicators for evaluating agricultural practices remain poorly understood and warrant further validation. Through field experiments, this study investigated the differential effects of agricultural practice treatments on soil properties and bacterial communities between two main farmland soil compartments: intra-row and inter-row. Additionally, we explored the potential correlations between key taxa and soil properties, as well as maize biomass. Results revealed marked disparities in soil properties, bacterial community compositions, and co-occurrence network patterns between intra-row and inter-row soils. Agricultural practice treatments exerted significant impacts on bacterial community structures and network topological features in both intra-row and inter-row soils. Subsequent correlation analysis demonstrated strong relationships between soil properties and most keystone species. In addition, 42 and 41 indicator species were identified in intra-row and inter-row soils, respectively, including shared genera such as Solirubrobacter, Blastococcus, Iamia, Conexibacter, and Lysobacter. Notably, 22 key indicator species in intra-row soils displayed significant positive/negative correlations with maize biomass, whereas only 4 key indicator species showed negative correlations in inter-row soils. These findings highlight differential responses of bacterial communities to agricultural practices in distinct soil compartments. The intra-row soils harbored more bacterial taxa significantly associated with maize biomass, while the inter-row soils better reflected the effects of agricultural interventions. This study confirms the spatial variability of microbial communities as effective bioindicators for evaluating agricultural practice strategies. Identification of compartment-specific indicators provides novel microbiological insights into supporting precision agriculture practices.

Keywords: agricultural practices; co-occurrence network; key indicator species; soil bacterial community; spatial heterogeneity.

Figure 1

Schematic diagram of the experimental design, field layout, and sample collection. (A) Layout of field experiments. Each individual plot is outlined in black lines. These plots, representing four treatments (yellow: Con_Mono, purple: Con_Inter, green: Fer_Mono, blue: Fer_Inter), were arranged semi-randomly in four blocks. (B) Schematics of Con_Mono (top) and Con_Inter (bottom). Green and yellow bands represent maize and soybean rows, respectively. Pink ovals outline approximate sampling areas in each plot. (C,D) Sampling sites and types for soil collection. Red ovals indicate sampling sites for intra-row and inter-row soils.

Figure 2

Soil properties between different treatments in intra-row and inter-row soils. Statistical significance of these differences was assessed using the Kruskal–Wallis test, and the results were presented at the top of each plot. The separate significance of differences between intra-row and inter-row soils under each treatment was assessed using the Wilcoxon rank sum test. ***: p < 0.001; **: p < 0.01; *: p < 0.05; ns: non-significant; TC: total carbon; TN: total nitrogen; C:N: total carbon/nitrogen ratio; DOC: dissolved organic carbon; NH₄⁺-N: ammonium nitrogen; NO₃⁻-N: nitrate nitrogen; MBC: microbial biomass carbon; MBN: microbial biomass nitrogen.

Figure 3

Distributions of the ASVs (A) and dominant phyla (B) of soil bacterial communities for each treatment in intra-row and inter-row soils. Phyla with relative abundances less than 1% were summarized as “Others”.

Figure 4

Relative abundances of dominant bacterial phyla in different treatments. Significance of differences between intra-row and inter-row soils was assessed using the Wilcoxon rank sum test. Results are presented at the bottom of each plot, with significant effects highlighted in bold. The separate significance of differences among treatments in intra-row and inter-row soils was assessed using the Kruskal–Wallis test. Different letters indicate significant differences among treatments.

Figure 5

Alpha diversity of soil bacterial communities between intra-row and inter-row soils. Significance of the differences was assessed using the Wilcoxon rank sum test. ***: p < 0.001.

Figure 6

General patterns of bacterial communities across soil samples. (A) PCoA shows bacterial community structures based on Bray–Curtis distance. Similarity of compartment and treatment was estimated via ANOSIM and presented at the top of the plot. Colored circles represent different compartment soils, while shapes represent the treatments, with 80% confidence ellipses shown around each compartment. (B) CAP shows the treatment patterns in intra-row and inter-row soils, respectively. The CAP analyses were constrained by “treatment”. The explained fraction of total variance is indicated above the plots. The percentage of variation shown on each axis refers to the explained fraction of total variance. Colored circles represent the four treatments, with 80% confidence ellipses shown around each treatment. (C) Bray–Curtis dissimilarity of bacterial communities among different treatments in intra-row and inter-row soils. The significance of differences among treatments was evaluated using the Kruskal–Wallis test. Different letters denote significant differences among treatments.

Figure 7

Co-occurrence patterns of bacterial communities among different treatments in intra-row and inter-row soils. (A,B) Co-occurrence networks for each treatment in intra-row (A) and inter-row (B) soils. Networks were constructed by calculating correlations among ASVs (|ρ| > 0.6, p < 0.001). Nodes are colored according to taxonomic classification at the phylum level. Colored edges represent positive (red) and negative (blue) correlations, respectively. Node sizes represent the degree of connections. Topological parameters (number of positive/negative edges, average degree, and clustering coefficient) are presented at the bottom of each plot. (C) Network robustness between intra-row and inter-row soils for each treatment. Results of linear models are displayed at the top of each plot. Colored circles represent intra-row (red) and inter-row soils (blue) soils, ***: p < 0.001.

Figure 8

Correlation networks of keystone species for different treatments in intra-row and inter-row soils. Keystone species with correlation coefficients greater than 0.9 were screened. Nodes are colored according to taxonomic classification at the phylum level. Node sizes represent the abundance of keystone species. Colored edges represent positive (red) and negative (blue) correlations.

Figure 9

Correlations between keystone species abundance and soil properties in intra-row (A) and inter-row (B) networks. The heatmaps present only the keystone species significantly related to soil properties. Positive (red) and negative (blue) correlations are indicated by color intensity. *: significant correlations.

Figure 10

Bipartite networks display indicator species in intra-row (A) and inter-row (B) soils. Specific ASVs significantly associated with one or more treatments were selected. Circles representing individual ASVs are colored according to their phylum assignments.