Comparative analysis of rhizobial and bacterial communities in experimental cotton fields: Impacts of conventional and conservation soil management in the Texas High Plains.

Conservative agricultural management strategies pursue long-term ecological benefits through practices such as no-tillage, cover crop, and inherent soil properties management. Farmers, however, are often hesitant to adopt such practices due to lack of experience, initial expense, and concern for low crop productivity. Overcoming this barrier requires novel approaches, such as effectively managing the soil microbiome to attain high productivity at a low cost, especially in a semi-arid region. To study the potential of conservation agriculture, we investigated components of soil bacterial community and rhizobial diversity in long-term experimental cotton fields divided into conventional tillage monoculture systems with winter fallow (CT) and no-tillage with mixed cover crop (M-NT) system on the Texas High Plain (THP). We conducted next-generation amplicon sequencing targeting rpoB gene with collected soil samples from different soil managements and seasons. Our research revealed that although CT had significantly greater bacterial diversity and species richness than the M-NT management, rhizobial diversity and species richness were higher in M-NT than in CT management. Both bacterial and rhizobial diversity and richness were greater in summer than in fall. The abundance of the order Rhizobiales was consistently high in M-NT than in CT fields in both seasons. Soil management altered the dominant rhizobial genus associated with cotton production systems; Rhizobium and Pararhizobium dominated M-NT management, while Bradyrhizobium and Sinorhizobium were dominant under CT management. These outcomes suggest that incorporating legumes into a cover crop in this semi-arid cotton-growing region can initiate beneficial changes to the dynamics of the indigenous rhizobial assemblage. The high prediction accuracy of our machine learning model using bacterial community data classifying the managements as CT or M-NT validates the possibilities of a strong underlying relationship between soil management and the bacterial diversity in the soil. • Overall soil bacterial diversity is higher under CT management. • Rhizobial diversity is higher in M-NT management. • Management practices alter bacterial and rhizobial assemblage composition. • The abundance of order Rhizobiales is higher under M-NT management. • Machine learning model recognizes management types with high prediction accuracy. [ABSTRACT FROM AUTHOR]