Personalized Crop Management for Sustainable Agriculture

Background/Question/Methods

While building long-term soil health is essential to producers from an ecological perspective, farmers are often inclined towards practices leading to short-term economic returns. Therefore, new approaches are required to address these two aspects simultaneously. Planting legumes as a cover crop has been a well-practiced and effective crop management strategy for centuries because of the symbiotic association between legumes and rhizobia that converts atmospheric nitrogen (N) to plant-available N. However, only certain rhizobia are compatible with a particular legume. In this context, our work had three objectives: (1) Evaluation of total bacterial diversity, and indigenous rhizobial assemblages of associated center-pivot irrigated conventional tillage with cotton monoculture (CT), and no-tillage with mixed species cover crop (Hairy vetch, Daikon radish, and winter pea)(M-NT) in experimental cotton fields of Texas High Plain (THP), a semi-arid region (2) Getting a list of the indigenous rhizobial pool (3) Investigating whether the microbiome data found in two types of fields can classify them through supervised machine learning. The soil sample was collected during the fall and summer seasons; the rpoB (RNA-Polymerase-Beta-Sub-unit) amplicon sequencing was used to identify total bacterial diversity and rhizobial assemblages to the species level.

Results/Conclusions

Our research revealed that the CT system had higher bacterial diversity and species richness than the M-NT, but rhizobial diversity and species richness was greater under the M-NT. CT system was dominated by genus Sinorhizobium, while Pararhizobium dominated M-NT. We also found that bacterial and rhizobial diversity was higher in summer than in fall. The high prediction accuracy of the machine learning model classifying the fields as CT or M-NT validates the underlying relationship between the farming strategy and the bacterial diversity in the soil. Pararhizobium giardinii has been identified as a potential biomarker for crop management strategies. The absence of rhizobia that is compatible with fairy vetch and winter pea for biological nitrogen fixation leaves room for improvement in crop management strategy. Incorporating external compatible rhizobia inoculum or planting legumes compatible with field-indigenous rhizobia may result in higher crop yield and better soil health. This study emphasizes the need to develop region-specific, personalized agricultural soil management