Nature is diverse but we still don't really know why. A number of mechanisms have been proposed to explain species coexistence and the diversity of natural systems. Deterministic control in the form of resource niches is a classical explanation. More recently, predators and pathogens, i.e. "top-down niches", have been identified as potentially important deterministic drivers of community structure. Furthermore, stochasticity could play a role under certain conditions such as dispersal limitation or a lack of resource limitation. The relative significance of these factors in natural communities is still unknown.
We studied the highly diverse natural communities of protists occurring in the aquatic habitats within bromeliad plants in Costa Rican rainforests and experimentally tested several proposed drivers of diversity. Specifically, we recorded a number of environmental variables and subsequently manipulated bottom-up control (resource supply), top-down control (predator presence) and dispersal on two spatial levels (within and between bromeliad metacommunities). We measured community change in terms of abundance, diversity and morphospecies composition.
Results/Conclusions
We found that protist abundance strongly increased with resource addition, while species richness (alpha diversity) was negatively affected. We also observed a clear shift in functional structure of protist communities with resource addition: flagellates, ciliates and predatory microfauna (copepods, rotifers) increased in abundance and species richness while amoebae and algae decreased. We show that not only was morphospecies composition significantly different in those communities with high resource availability compared to control communities, but also was there more variability in composition among those high-resource communities. Predator presence had only weak effects on certain groups (e.g. flagellates) and dispersal manipulation did not modify abundance, diversity or composition of protists.
Our results show that bromeliad protist communities might not be subject to top-down control or dispersal limitation. The latter mechanism is currently being tested in an additional colonization experiment. We conclude that protist communities in Costa Rican bromeliads are strongly bottom-up controlled and that removing resource limitation in these habitats can lead to highly stochastic assembly of protist communities with potentially far-reaching consequences for the rest of the bromeliad food web and its functions.