Responses of soil bacterial community structure and function to dry–wet cycles more stable in paddy than in dryland agricultural ecosystems.

Aim: Climate change is the foremost cause of terrestrial biodiversity decline, but the microbial communities from different habitats do not respond equally to environmental change. However, our understanding of the response patterns of microbial diversity to climate change remains limited in distinct agriculture ecosystems across large spatial scales. Here, we explore the response patterns of soil bacterial community structure and function to dry–wet cycles in dryland (maize) and paddy (rice) agricultural ecosystems throughout eastern China. Location: Eastern China. Time period: November 2019 to May 2020. Major taxa studied: Bacterial communities. Methods: Incubation experiments with dry–wet cycles were carried out on the soil samples collected from 25 paired maize and rice fields across eastern China. Bacterial community structure was estimated via high‐throughput sequencing techniques. We applied quantitative polymerase chain reaction to evaluate the effect of the dry–wet cycle on the change of functional genes. Phylogenetic signals were examined based on Blomberg's K to test the phylogenetic conservatism of maize and rice bacterial communities during dry–wet cycles. Results: Dry–wet cycles significantly decreased bacterial alpha diversity in both maize and rice soils, and bacterial communities from warmer regions were more sensitive to dry–wet cycles. The bacterial responses to dry–wet cycles were phylogenetically conserved, with greater phylogenetic conservation in the former. The abundance of functional genes that were related to carbon decomposition, nitrogen fixation, nitrification, denitrification and organophosphorus mineralization decreased during dry–wet cycles, and the resistances of such functional genes to disturbance were stronger in rice soils than in maize soils. In addition, the relative abundances of stable amplicon sequence variants during dry–wet cycles had a significant positive linear relationship with functional gene abundance in the maize and rice soils, indicating their essential roles in the stability of microbial function. Main conclusions: We conclude that soil bacterial community structure and function in dryland agricultural ecosystems are less stable and exhibit greater phylogenetic conservation in response to dry–wet cycles when compared with paddy fields. In addition, our study reveals, for the first time, the response patterns of bacterial communities to dry–wet cycles in dryland and wetland agroecosystems. [ABSTRACT FROM AUTHOR]