Open Collections

Stochastic and deterministic assembly within forest plant communities of British Columbia

University of British Columbia

Patterns of beta diversity, the variation in species composition among sites, are often used to gain insights into the processes governing plant community assembly. Deterministic processes including environmental selection and stochastic processes such as stochastic colonization and extinction, and / or priority effects vary in their relative importance for explaining patterns of diversity. Sampling strategies that simultaneously control for variation in the environment as well as gradients in species richness improve our ability to quantify the relative importance of stochastic and deterministic community assembly processes. Using data that control for environmental variation, collected from 809 standardized survey plots across British Columbia, Canada, I show using a null model analysis that controls for species richness gradients, that patterns of beta diversity are no different than expected based on random sampling within 31 site units sampled from the Interior Cedar Hemlock (ICH), Englemann Spruce Subalpine Fir (ESSF), Sub-Boreal Pine and Spruce (SBPS) and Boreal White and Black Spruce (BWBS) zones described by Biogeoclimatic Ecosystem Classification (BEC). I also show that deterministic assembly is revealed when beta diversity is compared to patterns expected under broader definitions of the species pools, but not consistently across plant lifeforms or across categories defined by BEC. I further found that dispersal limitation influences beta diversity and increases in importance from shrubs to herbs to trees. By using a null model approach that enables detection of stochastic assembly, I was also able to show that beta diversity is no different than expected based on random sampling for tree species composition within the subzone / variant level of BEC, which supports an underlying assumption of Bioclimate Envelope Models. I suggest that the high beta diversity within BEC site units is likely attributable to stochastic assembly processes, which deserve more attention in future research, especially at fine scales of community organization.

Text

2013-06-14

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/44566

Biology

University of British Columbia

UBCO

http://creativecommons.org/licenses/by-nc-nd/4.0/