Effects of human-induced stressors on microbial community landscapes in forested headwater wetlands: Are there clear links to process?

Background/Question/Methods

Previous studies have documented changes in the structure of wetlands impacted by human activities, with focuses on hydrological connectivity, vegetative community composition, and soil properties and their spatial variability.  While the spatial distribution of microbial communities is generally thought of as random variability, these communities can both be altered by and drive changes in these structural properties, but these links and feedbacks are not well understood.  Here we determined the relationship between soil properties and microbial communities, as measured by phospholipid fatty acid analysis (e.g., PLFA biomass, fungal-to-bacterial ratios and microbial composition), across a broad range of soil conditions found in headwater wetlands of Pennsylvania.  We ask if microbial communities and their spatial variability differ between wetlands classified as reference standard (n = 4) and those impacted by human activities (n = 4).  We also ask if the changes in microbial community measures at the site level reflect changes in site-specific carbon sequestration rates, calculated using Cesium-137 (¹³⁷Cs) based soil accretion rates and bulk density.

Results/Conclusions

Soil organic matter, soil texture and pH were among the best predictors of microbial community measures, with clear differences in both the abundances and composition of microbial communities between reference standard and impacted wetlands.  The differences found in microbial community abundance measures were strongly related to the lack of an organic horizon at human impacted sites, while differences in microbial composition were also related to differences within the mineral soils of reference standard and impacted sites.  These differences were related to variation in soil texture and pH. Using predictions from the statistical models and Bayesian kriging, we also found fine-scale and site-scale homogenization of the microbial community measures in impacted wetlands compared to reference standard wetlands.  Site-average total PLFA biomass was related to the average carbon content above ¹³⁷Cs peaks.  However, microbial community measures were not strongly related to carbon sequestration rates, where differences in hydrology are suspected to play a larger role in the mass transport of carbon into these systems.  These results highlight the complexity of soil properties and their potential influence on microbial community spatial patterning in headwater wetlands, as well as the challenges associated with linking changes in microbial communities to changes in wetland processes.