Influence of resource gradients on soil microbial communities of a polar desert

Microbial communities are major contributors to globally-significant ecosystem processes, yet our understanding of controls over bacterial community assembly and structure remains limited.  To gain a better understanding of the mechanisms important for shaping microbial community structure, here we apply macroecological models developed to describe productivity-diversity relationships in plant communities to soil bacterial communities of the McMurdo Dry Valleys, Antarctica. This work describes the influence of productivity gradients on soil bacterial diversity and metabolic activity in an energy-constrained environment. Soils of the Antarctic Dry Valleys exhibit strong natural gradients in properties such as organic matter content, water availability, and salt composition/concentration, all of which influence the structure of trophically simple communities.  Such properties make the Dry Valleys a model ecosystem for exploring relationships between biogeochemical and biological parameters. 

Soils of the McMurdo Dry Valleys were sampled in the austral summer of 2010 from sites representing a gradient of organic carbon and water content.  Soils were transported to Virginia Tech and analyzed for a variety of properties including microbial biomass, moisture content, soluble ion concentration, and soil organic matter concentration.  We assessed soil productivity using an index based upon extractable chlorophyll a content normalized by organic matter content. A terminal restriction fragment length polymorphism (T-RFLP) technique was used to assess community structure and diversity based on the 16S rRNA gene.  Potential microbial functioning was estimated through extra-cellular enzyme activities of two common organic matter degrading enzymes (alpha- and beta-glucosidase).

Results show that soil organic carbon ranged over an order of magnitude across the sampled productivity gradient, while productivity itself spanned nearly two orders of magnitude.  Significant variation in soil organic carbon, moisture content, total nitrogen, enzyme activity, and microbial biomass were also evident along the productivity gradient (ANOVA p-value ≤ 0.05).  PCA ordination of environmental data indicates separation of samples based strongly on these same soil properties (correlation with PC1 and PC2 axes; r > |0.6|).  In contrast ordinations of bacterial community TRFLP data suggest the separation of samples based primarily on soil physical characteristics, among them pH, which has been shown to be a major driver of soil bacterial diversity in soils worldwide.  Future work will use pyrosequencing technology to assess variation in microbial community composition that may result from resource availability and yet not be reflected in measures of community diversity examined here.