Dispersal limits rare species establishment in restored habitats more than competition or stress

Background/Question/Methods

Understanding the factors that limit species occupancy within restored habitats has been a longstanding goal of restoration ecology, particularly with respect to differences between rare and widespread species. One idea put forward to explain differences in habitat occupancy between rare and widespread species is that species tradeoff the ability to tolerate abiotic stress (e.g., low moisture requirements) for competitive ability. This in turn tends to relegate rare species to spatially or temporally rare environments with high stress and low competition. We sought to test this theory for a large suite of rare versus widespread species (n=32) in a restored habitat to assess the factors most limiting seedling establishment. Restored Missouri Ozark glades are highly fragmented prairie-like habitats dominated by competitive grasses, periodic fires, and summer droughts.  In a fully factorial experiment conducted under 45 rain-out shelters placed in a restored glade, we manipulated abiotic stress by simulating rainfall regimes typical of dry, average, and wet years and manipulated competition through burning, removing unseeded competitors, and unmanipulated controls. Seedlings were censused bimonthly for survivorship.  The effect of regional occupancy and treatment on percent survival and biomass were assessed using linear mixed-effects models.

Results/Conclusions

Results from this experiment showed no significant interactions between rarity and the competition or water stress treatments on percent seedling survival.  In the rainfall treatments, drier conditions reduced seedling survival equally for all species, regardless of regional occupancy. When competition was reduced via controlled burning or hand clipping, seedling survival for all species was relatively high in comparison to the control treatment.  Surprisingly, rare species performed as well as widespread species in all treatments and average seedling survival was relatively high (~50%).  Together, these results suggest that regional occupancy in this restored system is strongly limited by dispersal ability rather than competition or stress tolerance.  Future management practices for this highly fragmented system should consider increasing opportunities for rare species establishment through controlled burns and facilitated dispersal.