Drivers of soil bacterial community composition in agricultural systems of Alberta

The Soil Quality Monitoring Program (SQMP) was conducted from 1997 to 2007 and aimed to characterize agricultural soil quality from benchmark sites across Alberta by evaluating multiple soil physico-chemical parameters. However, no biological indicators of soil quality were included at that time. Given that agricultural practices can affect soil function and microbial communities are sensitive indicators of functional changes, we revisited the SQMP benchmark sites and added a survey of the bacterial community to the physico-chemical parameters. We evaluated the effect of agricultural practices (i.e., tillage intensity, crop type, herbicide use, and fertilization method), as determined by farmer surveys, on bacterial community composition, heterogeneity, diversity, and co-occurrence via high-throughput sequencing of the 16S rRNA marker gene. Agricultural practices influenced the heterogeneity and evenness of bacterial communities but did not play a major role in shaping their composition; instead, pH and ecoregion were identified as main drivers. Co-occurrence network analyses revealed that the complexity and behavior of interactions among members of the soil bacterial community are altered by different agricultural practices. Altogether, our results indicate that bacterial community composition is insensitive to different agricultural practices such as tillage, fertilization, crop type and herbicide usage in a one year survey. However, co-occurrence network metrics may be a promising indicator of soil health in agricultural systems monitored over time