Phylogenetic and functional responses to disturbance in mesic grasslands: A cross-continental comparison

Background/Question/Methods

The role of fire and grazing in structuring grassland communities has been recognized and widely studied.  But a key in understanding the generality of these responses globally is to conduct experimental studies of identical design in regions of varied biogeographic histories, i.e. speciation, extinction, dispersal, disturbance histories.   My study was conducted in the context of an ongoing long-term experiment with varied fire and grazing treatments at three mesic grassland sites in South Africa and North America: Kruger National Park, South Africa, Ukulinga Research Farm, Pietermaritzburg, South Africa and Konza Prairie, Kansas.  I examined phylogenetic and fuctional shifts in grass species, which constitute the bulk of the biomass in these grassland  communities, in response to altered grazing and fire regimes at these sites.  The most abundant grasses present at each site were collected, sequenced and a Bayesian phylogenetic tree was constructed using BEAST.   Functional traits hypothesized to relate to fire resistance, regeneration rates and palatability were measured for all dominant grasses across sites and treatments.   Significant shifts in species abundances between treatments were assessed using linear mixed models.   Clades exhibiting patterns of positive or negative responses to disturbance were identified using a recently developed node-based metric used to detect patterns of over and under-representation in phylogenetic communities.  Functional responses to fire and grazing were assessed using community aggregated trait value metrics and a node-based Rao’s entropy metric to detect concentrations of particular trait values in clades.  

Results/Conclusions

Despite no overlap in the identity of species between South Africa and North America, clades present in both regions generally exhibited a similar directional response to disturbance.  But the magnitude of the response to greater fire frequency in South Africa was much stronger than in North America, i.e.  the identity and abundances of the dominants exhibited more significant shifts in SA than in NA. These results suggest a general conservatism or evolutionary stasis in grasses phylogenetic response to fire and grazing, yet a greater sensitivity to fire in grasslands of SA.  In preliminary analyses of the functional responses to fire and grazing I found height and stomatal size to increase in response to more frequent fire and the absence of grazing.