Microbial colonization in recently exposed glacial forelands of the Antarctic Peninsula

Background/Question/Methods Along the west coast of the Antarctic Peninsula the vast majority of glaciers have retreated over the last 20 years as a result of global warming. Vascular plants have rapidly colonized some of the resulting newly exposed glacial forelands, implying rapid pedogenesis. We hypothesize that rapid microbial colonization and activity may help condition the recently exposed mineral substrate, thus facilitating the establishment of vegetation. We examined patterns in soil microbial populations as well as soil carbon (C) and nitrogen (N) cycling at one such newly exposed glacial foreland as a function of distance from the glacier. Rates of photosynthesis, respiration, aerobic ammonia oxidation (AAO), denitrification and N₂-fixation were measured, and molecular fingerprinting techniques were used to assess the microbial diversity.

Results/Conclusions Microbial photosynthesis rates ranged from 150 to 1300 µmol m-2 h-1. Respiration rates in the dark varied from 800-1800 µmol m-2 h-1, which exceed photosynthesis rates and suggest the presence of allochtonus organic carbon inputs. Some N transformations were also significant, particularly N₂-fixation (10-80 µmol m-2 h-1) and AAO (5-20 µmol m-2 h-1). Denitrification rates were < 3.6 µmol m-2 h-1. None of these biogeochemical transformation rates showed any significant trends with distance from the glacier. Microbial richness and diversity (based on 16S rRNA phylotypes) were similarly low compared to well-developed soils, and constant with distance from the glacier. Well-known forms of cold-loving cyanobacteria (such as Microcoleus antarcticus and Phormidium autumnale) were found to be the dominant phototrophs in these samples, but no N₂-fixing cyanobacteria could be detected, suggesting that heterotrophic N₂-fixation may be important. Because microbial colonization and activity were high, it remains a possibility that microbes contribute to soil conditioning in some way. However, because we could not detect spatial trends in communities, rates or most other soil parameters, a clear process of soil development could not be demonstrated.