Testing a Theoretical Model that Predicts Thresholds in Populations Forced by Imposing Random, Episodic Disturbances

We do not fully understand reasons behind extinction of populations and species. Consequently, our ability to anticipate extinction (which can be considered a permanent type of an ecological threshold) has remained elusive. In particular, it is not clear how the attributes of episodic disturbance regimes can elicit extinction. Here, I test a stochastic model that predicts population extinction based on attributes of the disturbance regime and population growth rates using phytoplankton in a test tube. I examined the response of phytoplankton (Thalassiosira weissflogii) to stochastic disturbances implemented by having MATLAB control a hydraulic pump that episodically removed portions of the population through time in between episodes of population recovery. My experiment demonstrated evidence that a theoretical or mathematical threshold predicted to exist in theory may actually apply to a diatom population dynamics in a test tube; however, my results also open new questions about how the statistical attributes of the disturbance regimes under study may alter the predicted time to population extinction.