Changes in land use driven by urbanization impact nitrogen cycling and the microbial community composition in soils

Abstract

Transition of populations from rural to urban living causes landscape changes and alters the functionality of soil ecosystems. It is unclear how this urbanization disturbs the microbial ecology of soils and how the disruption influences nitrogen cycling. In this study, microbial communities in turfgrass-grown soils from urban and suburban areas around Xiamen City were compared to microbial communities in the soils from rural farmlands. The potential N₂O emissions, potential denitrification activity, and abundances of denitrifiers were higher in the rural farmland soils compared with the turfgrass soils. Ammonia oxidizing archaea (AOA) were more abundant than ammonia oxidizing bacteria (AOB) in turfgrass soils. Within turfgrass soils, the potential nitrification activities and AOA abundances were higher in the urban than in the suburban soils. These results indicate a more pivotal role of AOA in nitrification, especially in urban soils. Microbial community composition was distinctly grouped along urbanization categories (urban, suburban, and rural) classified according to the population density, which can in part be attributed to the differences in soil properties. These observed changes could potentially have a broader impact on soil nutrient availability and greenhouse gas emissions.

Figure 1. Soil chemical properties in urban (blue), suburban (green) and rural (red) categories.

Error bars represent standard error of the mean (n = 32, n = 16 and n = 8 for urban, suburban and rural, respectively), and differences are significant when no same letter above the bars exists (p < 0.05). ND, not detected.

Figure 2

PDA (a), potential N₂O emission (b), N₂O/(N₂O + N₂) (c) and PNR (d) in urban (blue), suburban (green) and rural (red) categories. Error bars represent standard error of the mean (n = 32, n = 16 and n = 8 for urban, suburban and rural, respectively), and differences are significant when the letter above the bars are different (p < 0.05). PDA, potential denitrification activity; PNR, potential nitrification rate; N₂O/(N₂O + N₂), potential N₂O emission/PDA, the proportion of N₂O emission to the total gas emission in denitrification.

Figure 3. Heatmap showing average relative abundances of microbial communities at phylum level.

The clustering of taxa is based on the Pearson’s correlation.

Figure 4

(a) Canonical correspondence analysis (CCA) based on the relative abundance of each OTU. 16 significant soil properties and 2 urban data were selected as the environmental variables. The quantitative variables are represented by arrows and the categorical variable, land use (“Turfgrass” and “Farmland”), is shown as unfilled diamonds. (b) Variation partitioning analysis for the explanatory proportions of different sets of factors. The land use was transformed to dummy variables for this analysis (turfgrass, 1; farmland, 0).

Figure 5. Non-strict version of LEfSe results on microbial communities.

The cladogram indicates the taxa (highlighted with small circles and shading) showing different abundance values (according to LEfSe) in the urban, suburban, and rural agricultural soils. For each taxon (circle), the color denotes the significantly higher abundance of the taxon in the corresponding group. Yellow denotes that the taxon is not significantly higher in any group.