Ecological insights into soil health according to the life-history traits and environment-wide associations of bacteria in agricultural soils

Abstract

Soil health assessment may be enhanced by monitoring changes in bacterial populations that are indicators of various biological, physical, and chemical properties of soil. However, the lack of ecological information for many abundant bacteria in agricultural soils limits our understanding of indicator responses and, thus, their utility for guiding management. We identified bacterial indicators of twelve conventional measures of soil health, and tillage intensity, from a 16S rRNA gene-based survey of farmland across North America. We then analyzed trends according to bacterial life-history frameworks and an environment-wide association survey (EWAS) to gain ecological insights. Life-history traits were assessed using genomic traits inferred from taxonomic classifications and included: genome size, rrn copy number, and coding density. An EWAS was conducted using 89 studies of agricultural land management. Most bacterial indicators were positively correlated with biological measures and negatively correlated with physical and chemical measures of soil health, revealing broad differences in the way management shapes bacterial associations with soil health. High soil health ratings corresponded with life-history traits associated with metabolic dependency (smaller genome and lower coding density), while lower health scores corresponded with traits selected for by environmental instability and disturbance (larger genome and multiple rrn). Trade-offs in community-weighted genome size explained most variation in overall health score. EWAS confirmed the importance of disturbance-adapted bacterial indicators, underscoring the impacts of tillage on soil bacterial communities. These findings provide insights into the ecological relationships between bacterial indicators and soil health and illustrate new approaches for interpreting patterns in microbiome data.