Divergent responses of the native grassland soil microbiome to heavy grazing between spring and fall

Abstract

Grasslands are estimated to cover about 40% of the earth's land area and are primarily used for grazing. Despite their importance globally, there is a paucity of information on long-term grazing effects on the soil microbiome. We used a 68-year-old grazing experiment to determine differences in the soil permanganate-oxidizable C (POXC), microbial biomass C (MBC), the soil prokaryotic (bacterial and archaeal) community composition and enzyme activities between no-grazing, light grazing and heavy grazing, i.e. 0, 1.2 and 2.4 animal unit months (AUM) ha^-1. The grazing effects were determined in spring and fall grazing. Light grazing had little effect on soil MBC and the composition and diversity of prokaryotic communities in either grazing season, but the effects of heavy grazing depended on the grazing season. In spring, heavy grazing increased the relative abundances of copiotrophic phyla Actinomycetota, Bacillota and Nitrososphaerota, along with soil POXC contents but decreased those of oligotrophic phyla Acidobacteriota, Verrucomicrobiota and Nitrospirota. This difference in responses was not observed in fall, when grazing reduced soil POXC, MBC and the relative abundances of most phyla. The β-diversity analysis showed that the prokaryotic community structure under heavy grazing was different from those in the control and light grazing treatments, and α-diversity indices (except the Shannon index) were highest under heavy grazing in both grazing seasons. The activities of P-mobilizing and S-mobilizing soil enzymes decreased with increasing cattle stocking rate in both seasons, but the activities of the enzymes that mediate C and N cycling decreased only in the fall. The genus RB41 (phylum Acidobacteriota) was one of two core bacterial genera, and its relative abundance was positively correlated with the activity of the S-mobilizing enzyme. Therefore, light grazing is recommended to reduce negative effects on the grassland soil microbiome and its activity, and the grazing season should be considered when evaluating such grazing effects.

Keywords: cattle stocking rate; nutrient cycling; soil microbial diversity.

Fig. 1.. Soil POXC and MBC at different cattle stocking rates in spring and fall.*p<0.05, **p<0.01. NS, not significant.

Fig. 2.. The relative abundances of soil prokaryotic (bacterial and archaeal) phyla at different cattle stocking rates in spring and fall. NS. not significant. *p<0.05, **p<0.01.

Fig. 3.. PCoA showing the cattle stocking impacts on the β-diversity of the soil prokaryotic communities (a) (PERMANOVA F-value = 3.4345, r² = 0.10755 and P = 0.001) and the differential abundances of the prokaryotic genera (determined by LEfSe) at different stocking rates (b). LDA, linear discriminant analysis score in LEfSe.

Fig. 4.. PCoA showing the grazing season impacts on the β-diversity of the soil prokaryotic communities (a) (PERMANOVA F-value = 2.5588, r²=0.042252 and P=0.001) and the differential abundances of the prokaryotic genera (determined by LEfSe) between the two seasons (b). LDA, linear discriminant analysis score in LEfSe.

Fig. 5.. The core prokaryotic genera that were observed in 85% of the soil samples at all cattle stocking rates and both grazing seasons.

Fig. 6.. Soil enzyme activities at different cattle stocking rates in spring and fall. NS, not significant. *p<0.05, **p<0.01.

Fig. 7.. Relationships between arylsulphatase enzyme activities and the relative abundances of the genera RB41 (a) and Mycobacterium (b).