COS 27-7 - It’s all relative: Bacterial richness varies across global habitats

Tuesday, August 8, 2017: 10:10 AM
B116, Oregon Convention Center
Kendra E. Walters and Jennifer B.H. Martiny, Ecology and Evolutionary Biology, University of California, Irvine, Irvine, CA

Broad-scale biogeographic patterns well documented for plants and animals remain uncharacterized for bacteria and other microorganisms. Even though measures of diversity are often used to compare microbial communities within a single study, it is still unclear which ecosystems and habitats might harbor the richest bacterial communities. Often problematic are differences in OTU classifications and sequencing depth among studies that make direct comparisons of diversity challenging. However, given the availability of databases that house datasets large enough and with deep enough sequencing, relative richness can now be assessed among a global range of ecosystems and habitats. In this study, we quantified bacterial richness across eleven ecosystems and sixteen habitats using data from the Earth Microbiome Project (EMP). These data encompass 95 studies, totaling 22,112 samples, from all seven continents. The EMP uses the 16S rRNA region to classify all sequences into OTUs at 97% sequence similarity with a median depth of 54,091 sequences per sample, ensuring the comparability among samples. The samples were rarefied to 15,000 sequences, and a suite of alpha diversity indices calculated for each. Ecosystem and habitat classifications were curated for each sample based on accompanying metadata, and median alpha diversity metrics were calculated across 1000 replicate rarefactions.


Observed bacterial richness (OTUs) appeared highest in soils and sediments, habitats characterized by high heterogeneity. Within soils, high diversity was found in ecosystems with higher energy/nutrient availability (agriculture, grasslands, shrublands, hot deserts, and forests) than ecosystems of low soil diversity (tundra and cold deserts). In contrast, lower bacterial richness was observed in water and air, habitats of low heterogeneity, and host-associated habitats. Overall, the global biodiversity patterns uncovered here are consistent with patterns found in other biodiversity studies of a smaller scale, and they indicate that bacterial diversity may be controlled by similar environmental factors that control plant and animal diversity: habitat heterogeneity and energy availability. We are currently comparing results based on observed OTU richness to other alpha diversity indices such as evenness or dominance. Future directions include analyzing global patterns of beta diversity to shed light on the surprisingly high richness observed in low plant and animal diversity ecosystems, such as agricultural sites and hot deserts. This study provides an overview of global bacterial diversity that can serve as a baseline for comparison with other studies and provides insight into how habitat heterogeneity and energy availability may be impacting bacterial biodiversity.