Species driving food web stability have lower individual variation

Background/Question/Methods

Individual variation can potentially affect the dynamics of populations and communities through several mechanisms. One such mechanism relates trait variation to dietary variation, which may in turn affect food web structure and stability. The extent to which this effect occurs is largely unknown, and is the focus of the present study. It has been proposed that diet variation can be linked to the number and strength of interactions of a given species in a food web. The larger dietary variation is, the larger the number of interactions, but the weaker each of those interactions become and vice versa. However, species-level dietary variation can be decomposed into within-individual variation and between-individual variation. Higher within and between-individual variation leads to generalist species, while lower within or between-individual variation leads to specialists, but a large range of intermediate cases are possible. Here, I take the latter into account and I assess how individual variation affects the chance that a given species drives food web stability, using a combination of May’s classical approach, and simulated results on dominant eigenvectors as a proxy for species idiosyncratic effects. I also derive an analytic expression relating variation and the idiosyncratic effect of species upon stability.

Results/Conclusions

I analytically predicted that low levels of individual variation in diet should lead to species with the highest likelihood of determining the stability of randomly assembled food webs, either increasing or decreasing it. My simulations support these results, and further suggest that the chance of a given species to determine stability is severely reduced in randomly assembled food webs whenever individual variation is high. However, species with high levels of individual variation can still have a strong impact upon food web stability whenever they have a similar number of preys and predators. Finally, I found that individual variation mainly affects stability through its effect upon interaction strength in randomly assembled food webs. Nevertheless, the effect of individual variation in stability through connectivity may be more important in food webs with realistic structure. We thus need to take into account more realistic food web structures to fully understand the effect of individual variability upon stability. These results suggest that individual variation can have strong impacts upon food web structure and stability, and underlines the need for comprehensive studies that take into account trait and dietary variation.