COS 105-5
Population size affects competitive outcomes in a model protist system via stochastic drift

Thursday, August 14, 2014: 9:20 AM
Beavis, Sheraton Hotel
Geoffrey Legault, Ecology and Evolutionary Biology, University of Colorado at Boulder, Boulder, CO
Jeremy W. Fox, Dept. of Biological Sciences, University of Calgary, Calgary, AB, Canada
Brett A. Melbourne, Department of Ecology and Evolutionary Biology, University of Colorado at Boulder, Boulder, CO
Background/Question/Methods

           Populations are the outcomes of individual demographic events such as birth, death, and migration. In nature, these events are inherently probabilistic. The implications of this demographic stochasticity vary depending on whether populations are large or small. Since large populations are the result of many individual demographic events, their dynamics should be close to the expected value of the stochastic process (or a deterministic equivalent). As populations get smaller and events less numerous, however, outcomes may start to deviate more markedly from the expected value. Population size is therefore an important determinant of how the inherent stochasticity of demographic events affects the predictability of population dynamics. To date, this effect has rarely been tested directly, particularly for multi-species assemblages. Given that many populations are declining due to habitat loss, it is important to better understand the size-predictability relationship in real populations of interacting species.

We sought to test this relationship empirically by controlling the population size of single-celled protists over many generations. To do so, we created small (1 mL), medium (10 mL), and large (80 mL) communities containing combinations of one, two, and three species of bacterivorous protists (12 replicates each).  We monitored the dynamics of the single-species communities for 42 days (> 50 generations), and the dynamics of the multi-species communities until competitive exclusion had occurred (up to 60 days). State-space stochastic growth/competition models were fitted to the data to assess whether differences in the predictability of replicate populations could be attributed to the size-related effects of demographic stochasticity.

Results/Conclusions

           In line with theory, variability between replicates increased as population size decreased for both the single- and multi-species combinations. This meant that smaller population sizes led to larger between replicate differences in the observed growth rate, carrying capacity, and the onset of growth/decline. Further, smaller multi-species communities had greater variation in terms of both the timing of competitive exclusion and the identity of the winner. These results highlight how the effects of demographic stochasticity can contribute to unpredictable outcomes when populations are small. This implies that even strong competitors in nature may be at risk of losing out as populations shrink and outcomes become more variable.