Anatomy of biodiversity and community structure change on a human land use gradient

Background/Question/Methods

The human-driven crisis of worldwide biodiversity loss intensifies the need to better understand how biodiversity and human land use are related.  We lack an adequate understanding of how ecological communities change in composition and structure with human land use.   We determine primary community metrics for boreal vascular plants with two aims: i) to assess the land use associated risks facing individual communities and the broader regional metacommunity, and ii) to determine the utility of metrics and their sensitivity to human land use for monitoring purposes. 

We sampled vascular plant species occupancy along a 0 – 100% continuum of human land use footprint in the 380 000 km² boreal ecoregion of northern Alberta, Canada.  In each of 320 sites selected systematically throughout the region, we determined species occupancy within a one hectare area, and human footprint extent at within a 150 m radius. 

Specifically, we compare how homogeneity, species commonness/rarity, and specialization, relate to disturbance using rank species occupancy relationships and species subsets in richness-disturbance relationships.  Because disturbed communities are also often expected to contain species with more similar traits and more similar evolutionary heritage, we also compare the phylogenetic structure and phenotypic functional trait structure across communities with varying human land use.

Results/Conclusions

We observed a significant humped relationship between species richness and human land use, but this pattern was not as sensitive to land use as most other metrics of community structure. Contrary to expectations, communities of intermediate human land use (not those of greatest land use) exhibited greatest dominance by common species (with rare species least frequently observed).  Similarly, generalist species were more ubiquitous (and specialists less ubiquitous) with intermediate land use than areas of very high or low land use.  Generally communities were most homogeneous (similar to each other) in species composition when land use was of intermediate extent. These results contrast other studies of community patterns along disturbance gradients, but are consistent with extensions of the intermediate disturbance hypothesis.

Land use was not generally related to phylogenetic structure, but communities were composed of more functionally similar species with extensive human land use.  We suggest the observed phenotypic trait structuring in impacted communities did not translate to phylogenetic structure because phylogenetic structuring depends on trait conservatism. 

Here, species specialism and rarity exhibited the strongest responses to disturbance, and should be considered foci of biodiversity monitoring in this region.