Differential roles of deterministic and stochastic processes in structuring soil bacterial ecotypes across terrestrial ecosystems

Abstract

Soil bacteria are vital to ecosystem resilience and resistance, yet ecological attributes and the drivers governing their composition and distribution, especially for taxa varying in ecological traits and inhabiting different ecosystems, are not fully understood. Here, we analyzed a large-scale bacterial community and environmental dataset of 622 soil samples systematically collected by us from six major terrestrial ecosystems across the United States. We show that soil bacterial diversity and composition significantly differ among ecotypes and ecosystems, partially determined by a few universal abiotic factors (e.g., soil pH, calcium, and aluminum) and several ecotype- or ecosystem-specific ecological drivers. Co-occurrence network analysis suggests that rare taxa have stronger ecological relevance to the community than abundant taxa. Ecological models revealed that deterministic processes shape assembly of abundant taxa and generalists, while stochastic processes played a greater role in rare taxa and specialists. Also, bacterial communities in the shrubland ecosystem appear to be more sensitive to environmental changes than other ecosystems, evidenced by the lowest diversity, least connected community network, and strongest local environmental selection driven by surrounding land use. Overall, this study reveals ecological mechanisms underlying the bacterial biogeography in terrestrial ecosystems nationwide and highlights the need to preserve rare biosphere and shrubland ecosystems amid environmental disturbance.