Linkage of plant and abiotic properties to the abundance and activity of N-cycling microbial communities in Tibetan permafrost-affected regions

作  者:Chen YL, Kou D, Li F, Ding JZ, Yang GB, Fang K, Yang YH*
影响因子:3.306
刊物名称:Plant and Soil
出版年份:2019
卷:434  期:  页码:453–466

论文摘要:

Aims

Ammonia oxidation and denitrification are crucial for nitrogen (N) availability and nitrous oxide production in N-limited permafrost soils. However, it remains unclear about the relative roles of abiotic and biotic properties in controlling the abundance and activity of ammonia-oxidizing and denitrifying microorganisms in permafrost-affected soils.

Methods

We analysed the potential ammonia oxidation and denitrification rates (PAO and PDR), the abundance of archaeal amoA, bacterial amoAnirK, nirS and nosZ genes, soil characteristics, climatic and plant properties across two vegetation types in Tibetan permafrost-affected soils. The relative importance of abiotic and biotic properties in driving functional N gene abundance, PAO and PDR were assessed using variation partition analysis (VPA) and random forest (RF) model.

Results

The functional N gene abundance and PDR were lower in alpine steppe than in alpine meadow. Variations in the PAO and PDR and functional N gene abundance were mainly explained by abiotic variables such as organic carbon and total N, then by plant properties such as plant N concentration, plant species richness and productivity based on the VPA. The RF model showed that abiotic properties (e.g., precipitation) and plant properties (e.g., plant N concentration or plant productivity) predicted the PDR and the abundance of functional N genes. Both VPA and RF model showed that the PAO and PDR could be determined by the abundance of functional N genes such as archaeal amoA gene and nosZ gene, respectively.

Conclusions

Our study highlights that abiotic and plant properties are important predictors of the abundance and activity of ammonia-oxidizing and denitrifying communities in permafrost-affected regions, implying that plant properties, which were previously overlooked, should be incorporated into ecosystem models for improved prediction of belowground N process rates in a changing environment.