Search

Free‐living N2 fixation (FNF) is an important source of bioavailable nitrogen for forests. Though increasing plant species diversity (PSD) benefits soil nitrogen accumulation, how it impacts FNF rate has not been explored.Forty‐five plots covering a PSD gradient were selected in a subtropical karst forest, southwest China. FNF rates in leaf litter and soil were measured using acetylene reduction assay calibrated with a 15N2 fixation method. Diazotrophic community, litter and soil properties were determined as well.The nifH gene abundance was not significantly altered in litter but was significantly decreased in soil by higher PSD. Diazotrophic community's Shannon diversity was significantly increased by higher PSD in both litter and soil. Increasing PSD inhibited litter FNF rate via aggravating phosphorus limitation and decreasing the relative abundance of Betaproteobacteria and Deltaproteobacteria. For the mineral soil horizon, increasing PSD suppressed FNF rate via decreasing nifH gene abundance caused by elevated N availability and lowered phosphorus and iron availability or through decreasing the relative abundance of Deltaproteobacteria but increasing that of Betaproteobacteria.Our findings, for the first time, reveal the microbial and abiotic controls on FNF in litter and soil in response to PSD, and hence benefit the prediction of nitrogen input via biological N2 fixation under changes in PSD. Read the free Plain Language Summary for this article on the Journal blog. [ABSTRACT FROM AUTHOR]

Biological N2 fixation (BNF) is a major pathway of external nitrogen input to terrestrial ecosystems. Nevertheless, the relative effects of diazotrophic community and soil properties on symbiotic (SNF) or free-living (FNF) N2 fixation remain poorly understood. Here, 20 sites of kudzu (Pueraria lobata) communities, which were at the early succession stage following agricultural abandonment, were selected across a karst area in southwest China. BNF rates were determined using the acetylene reduction assay calibrated with the 15N2 fixation method. The average rate was 17.81 ± 2.17 mg N g−1 day−1 nodule for SNF and 45.36 ± 5.15 ng N g−1 day−1 soil for FNF. Diazotrophic communities were dominated by Bradyrhizobium at the genus level in both nodule and soil. According to structural equation modeling, the relative abundances of two diazotrophic genera (Desulfovibrio and Geobacter) were the strongest explanatory variable for the variation of SNF rates with the second strongest variable being nifH gene abundance. Soil water content and available phosphorus indirectly affected SNF via their effects on diazotrophic community composition. In contrast, soil FNF rates were most pronouncedly affected by the availability of vanadium and iron, followed by soil water content, available phosphorus, and the relative abundances of two diazotrophic genera, i.e., Burkholderia and Cupriavidus. Our findings therefore help to improve the understanding of the relationship between BNF rates and their associated diazotrophic communities, and hence benefit a better prediction of BNF under global change. [ABSTRACT FROM AUTHOR]