Tillage Promotes the Migration and Coexistence of Bacteria Communities from an Agro-Pastoral Ecotone of Tibet

Abstract

In the Tibetan agro-pastoral ecotone, which has an altitude of 4000 m above sea level, small-scale cropland tillage has been exploited on the grassland surrounding the houses of farmers and herdsmen. However, knowledge of the effects of land change from grassland to cropland on soil nutrients and microbial communities is poor. Here, we investigated the structure and assembly mechanism of bacterial communities in cropland (tillage) and grassland (non-tillage) from an agro-pastoral ecotone of Tibet. Results indicated that soil nutrients and composition of bacterial communities changed dramatically in the process of land-use change from grassland to cropland. The pH value and the content of total nitrogen, organic material, total potassium, and total phosphorus in cropland soil were well above those in grassland soil, whereas the soil bulk density and ammonia nitrogen content in grassland soil were higher than those in cropland soil. Proteobacteria (30.5%) and Acidobacteria (21.7%) were the key components in cropland soil, whereas Proteobacteria (31.5%) and Actinobacteria (27.7%) were the main components in grassland soils. Tillage promotes uniformity of bacterial communities in cropland soils. In particular, the higher migration rate may increase the coexistence patterns of the bacterial community in cropland soils. These results also suggest that the tillage promotes the migration and coexistence of bacterial communities in the grassland soil of an agro-pastoral ecotone. In addition, the stochastic process was the dominant assembly pattern of the bacterial community in cropland, whereas, in grassland soil, the community assembly was more deterministic. These findings provide new insights into the changes in soil nutrients and microbial communities during the conversion of grassland to cropland in the agro-pastoral ecotone.

Keywords: agro-pastoral region; bacterial community; coexistence; soil nutrients; tillage.

Figure 1

The difference in soil nutrients between cropland and grassland via independent-samples t-test. (A) The difference in soil total nitrogen between the cropland and grassland. (B) The difference in soil ammonia nitrogen (NH₄-N) between the cropland and grassland. (C) The difference in soil nitrate-nitrogen (NO₃-N) between the cropland and grassland. (D) The difference in soil organic matter between the cropland and grassland. (E) The difference in soil total potassium between the cropland and grassland. (F) The difference in soil available potassium between the cropland and grassland. (G) The difference in soil total phosphorus between the cropland and grassland. (H) The difference in soil available phosphorus between the cropland and grassland. ** shows the significant difference at the 0.01 level, and * shows the significant difference at the 0.05 level.

Figure 2

Relative abundance of main bacterial composition in the cropland and grassland soils at the phylum level.

Figure 3

Mantel test of environmental variables with their corresponding composition and alpha diversity of bacterial communities in cropland (A) and grassland soils (B). The list of abbreviations is as follows: soil bulk density (SBD), soil water content (SWC), pH, total nitrogen (TN), ammonia nitrogen (NH₄-N), nitrate-nitrogen (NO₃-N), total phosphorus (TP), available phosphorus (AP), total potassium (TK), available potassium (AK), and organic matter (OM). *** shows the significant difference at the 0.01 level, ** shows the significant difference at the 0.01 level, and * shows the significant difference at the 0.05 level.

Figure 4

The composition of persistent bacteria in cropland (A) and grassland soils (B), and across all cropland and grassland soils (C).

Figure 5

Assembly processes of bacterial communities in cropland and grassland soil from an agro-pastoral ecotone of Tibet. (A) Normalized stochasticity ratio (NST) of bacterial communities in cropland and grassland. A neutral community model was used to determine the contribution of stochastic processes to microbial community assembly in cropland (B) and grassland (C). In the model, “m” is the estimated migration rate, and the higher the value of “m”, the lower the degree of diffusion limitation of the bacterial community.

Figure 6

Co-occurrence networks of bacterial communities at the OUT level in the soil of cropland (A) and grassland (B) based on a Spearman correlation (|r| > 0.7, p < 0.05). The percentage of dominant bacteria in the networks (C). The difference in betweenness centrality of the network between cropland and grassland (D).

Figure 7

The conceptual model reveals the differences in the composition and assembly patterns of soil nutrients and bacterial communities in cropland and grassland. Green shows a decreasing trend from dark to light.