Soil biodiversity and ecosystem function: Insights from Antarctic low diversity soils

Background/Question/Methods

The biodiversity-ecosystem function (BEF) relationship has received much attention over the last few decades but there is still disagreement over the relative influence of species richness and species identity/composition on processes in belowground systems. Most studies that have investigated the BEF relationship in soils have used artificial diversity gradients. For example, some experiments have created diversity gradients by removing the biotic communities using fumigation or heating, after which the soils were inoculated with biota to create a diversity gradient. In other studies the soils were simply diluted to create a diversity gradient. In both cases the created communities are not natural, and this may have influenced the outcome of the experiment. It would be more appropriate to use soils that represent a natural biodiversity gradient. This can be achieved using Antarctic soils that are naturally species poor and comparing these with soils of higher known diversity. Hence, we compared the decomposition potential, a key ecosystem process, of Antarctic soils (n=24) representing a diversity gradient with that of more diverse soils from short-grass steppe (n=2) and aspen forest (n=1) in Colorado. Soil decomposition potentials were measured as percent mass loss of filter paper in soils during incubation at 15ºC.

Results/Conclusions

Pyrosequencing indicated that the 27 soils represented a natural diversity gradient. Prokaryotic species richness increased from Antarctic soils to short-grass steppe to Aspen forest, but there was great variation in species richness between Antarctic soils. The Antarctic soils also varied considerably in their decomposition potential and some soils showed no measurable decomposition. It is however noteworthy that a few of the Antarctic soils had decomposition potentials comparable to those found for the short-grass steppe soils. We found a very strong linear relationship between species richness and decomposition rates across all sites as well as a significant relationship between community composition and decomposition rates. Overall, we could explain more of the variation in decomposition using species richness measures than composition suggesting that, at least at low species richness levels, the number of species have a greater influence than community composition. Moreover, the relationship between number of OTUs and decomposition rates was stronger than the relationship between decomposition rates and any of the abiotic factors measured.