COS 7-4 - Vegetation community assembly patterns in boreal peatlands: using a spatial hierarchical approach

Background/Question/Methods

In the remote area of Eeyou Istchee James-Bay in north-western Québec, Canada, there is an important need to better understand vegetation communities and how they may interact with climate change and increasing human activities (mainly mining and hydroelectricity production). Since peatlands occupy 14-50% of the territory, we chose 36 undisturbed peatlands to study vegetation community assembly patterns across a 1000km gradient. For this, we developed an innovative spatial-hierarchical approach combining classical ordination methods, indicator species analysis, and randomisation techniques to test assembly hypotheses at various spatial scales (territory, region, landscape, and site) and across different taxonomical groups (tracheophytes, bryophytes sensu lato, and lichens).

Results/Conclusions

Ordinations showed that Muskeg communities dominated by mature Black Spruce can be found across the entire territory. More open peatland communities, while difficult to characterise because of their composition differences, were organised into two groups following a latitudinal gradient. Surprisingly, environmental factors did not prove to be important for assembling communities at the regional scale. Results suggested instead either stochastic processes or an interaction between opposing factors (competition and environmental filtering on one side and facilitation and dispersal on the other). Coarser grain factors (e.g., climate and physiography) at the territorial scale did prove important for distinguishing regional species pools of tracheophytes, sphagna, and lichens, but not bryophytes sensu stricto and liverworts. At the finest scale, biotic factors (e.g., facilitation) also proved significant. While contributing to baseline knowledge of peatland communities in the territory, our results show that different taxonomical groups respond differently to scale and that environmental factors have little effect or their effects may be integrated with other scales and factors. This understanding of how different communities and taxonomical groups respond to various assembly factors is necessary for better understanding resistance-resilience capacities of these communities over the near future.