Microorganisms play fundamental roles in the diversity and functional stability of environments, including nutrient and energy cycling. However, microbial biodiversity loss and change because of global climate and land use change remain poorly understood. Many microbial taxa exhibit fast growth rates and are highly sensitive to environmental change. This suggests they have potential to be efficient biological indicators to assess and monitor the state of the habitats within which they occur. Here, we describe and illustrate a range of univariate and multivariate statistical approaches that can be used to identify effective microbial indicators of environmental perturbations and quantify changes in microbial communities. We show that the integration of multiple approaches, such as linear discriminant analysis effect size and indicator value analysis, is optimal for the quantification of the effects of perturbation on microbial communities. We demonstrate the most prevalent techniques using microbial community data derived from soils under different land uses. We discuss the limitations to the development and use of microbial bioindicators and identify future research directions, such as the creation of reliable, standardised reference databases to provide baseline metrics that are indicative of healthy microbial communities. If reliable and globally-relevant microbial indicators of environmental health can be developed, there is enormous potential for their use, both as a standalone monitoring tool and via their integration with existing physical, chemical and biological measures of environmental health.
Keywords: Ecosystem health; Microbial indicators; Molecular; Statistics.