A unifying ecological theory of microbial biodiversity

An ecological theory of microbial biodiversity has yet to be developed. This shortcoming leaves patterns of abundance, distribution, and diversity for the most abundant and diverse organisms on Earth without a predictive framework. However, because of their high abundance and complex dynamics, microbial communities may be underpinned by lognormal dynamics, i.e., synergistic interactions among complex stochastic variables. Using a global-scale compilation of 20,456 sites from a diverse set of natural and host-related environments, we test whether a lognormal model predicts microbial distributions of abundance and diversity-abundance scaling laws better than other well-known models, including the most successful macroecological theory of biodiversity, i.e., maximum entropy theory of ecology. We found that the lognormal explains the greatest percent variation in abundance, that the success of the lognormal increased with abundance while other models decreased, and that the lognormal was the only model to reproduce recently documented diversity-abundance scaling laws. Our unifying ecological theory of microbial biodiversity explains and predicts macroecological patterns based on dynamics that capture the complex large number dynamics of microbial life.