Background/Question/Methods Temporal fluctuations in selection (or fitness differences) among asexual genotypes (or species) can help maintain diversity in natural systems. Since selection is estimated from changes in relative abundance, it can be reliably estimated from time-series data on abundance. With sufficient sampling effort, it is often possible to detect statistically significant variation, and thereby implicating it as a factor for the maintenance of diversity. However, one challenge that remains is to move beyond statistical significance and assess the biological significance of fluctuating selection. Stated another way, how might the diversity of a particular system respond if fluctuating selection was removed? Here we use a well-studied competition model to develop a time-series probe capable of elucidating the biological significance of fluctuation selection, and apply the probe to analysis of fluctuating selection in natural zooplankton populations.

Results/Conclusions Using the consumer-resource model, we studied the biological significance of fluctuating selection across a range of fluctuating dynamics from limit cycle to chaotic fluctuations. Interestingly, we found that fluctuating selection is more important under limit cycles than under chaotic dynamics. Our analysis of the model reveals that the amount of observed selection provides an upper bound to selection in the absence of any temporal variation. By applying the probe to 26 data sets of zooplankton genotypes, we find that despite statistically significant levels of fluctuating selection, the biological significance of fluctuating selection in these populations is surprisingly small.