Source-sink dynamics in a temporally heterogeneous environment

In traditional source-sink models, vital rates and movement probabilities are assumed to be temporally homogeneous. Numerous studies, however, have demonstrated that temporal heterogeneity is the rule rather than the exception in natural systems. A case of particular interest is an expanding and contracting population where determination of source and sink populations is dependent on the timing and duration at which population growth rates are measured. Thus, ecologists have been perplexed about what is the proper scale to test for source-sink dynamics. In this study, I present a temporally variable source-sink dynamic driven by flooding disturbance. I measured population growth rates of an herbivorous neotropical rolled-leaf beetle, Cephaloleia fenestrata (Chrysomelidae), at different temporal scales in this expanding and contracting population. I demonstrate that flooding created a strong sink, but the system lacked source-sink dynamics during nonflood periods. The sink is caused by a flood-related sevenfold decrease in survival probability. There was no evidence for effects of flooding on recruitment. Migration was directional from the upland to the flood zone habitat, supporting the conclusion that the population in the flood zone is a sink. A Monte Carlo simulation demonstrated that whether the population in the flood zone is a sink, and the intensity of the sink, ultimately depend on the flooding frequency. Given an observed flood frequency of approximately once per year, the flood zone population was determined to be a long-term sink. This study provides evidence of high temporal variability resulting in fluctuations in and out of a source-sink dynamic and implicates a causal factor (increased mortality due to flooding). These results illustrate the importance of considering temporal variability in source-sink dynamics, and in choosing the proper temporal scale at which to test for source-sink dynamics.