The multiple meanings of omnivory influence empirical, modular theory and whole food web stability relationships

Abstract

The persistence of whole communities hinges on the presence of select interactions which act to stabilize communities making the identification of these keystone interactions critical. One potential candidate is omnivory, yet theoretical research on omnivory thus far has been dominated by a modular theory approach whereby an omnivore and consumer compete for a shared resource. Empirical research, however, has highlighted the presence of a broader suite of omnivory modules. Here, we integrate empirical data analysis and mathematical models to explore the influence of both omnivory module (including classic, multi‐resource, higher level, mutual predation and cannibalism) and omnivore–resource interaction type on food web stability. We use six classic empirical food webs to examine the prevalence of the different types of omnivory, a multi‐species consumer–resource model to determine the stability of these different kinds of omnivory within a module context, and finally extend these models to a 50 species, whole food web model to examine the influence of omnivory on whole food web persistence. Our results challenge the concept that omnivory is broadly stabilizing. In particular, we demonstrate that the impact of omnivory depends on the type of omnivory being examined with multi‐resource omnivory having the largest correlation with whole food web persistence. Moreover, our results highlight that we need to exercise caution when scaling modular theory to whole food web theory. Cannibalism, for example, was the most persistent and stable omnivory module in the modular theory analysis, but only demonstrated a weak correlation with whole food web persistence. Lastly, our results demonstrate that the frequency of omnivory modules are more important for whole food web persistence than the frequency of omnivore–resource interactions. Together, these results demonstrate that the role of omnivory often depends both on the type of omnivory being examined and the food web within which it is nested. In whole food web models, omnivory acts less as a keystone interaction, rather, specific types of omnivory, particularly multi‐resource omnivory, act as keystone modules. Future work integrating module and whole food web theory is critical for resolving the role of key interactions in food webs.

Disturbances in food webs : Importance of species interactions from an ecological and evolutionary perspective

“…Previous studies demonstrate that the network-level effect of omnivory on stability is highly context-dependent (Bascompte et al 2005;Monteiro et al 2016;McLeod et al 2021). The stability of omnivory motifs in isolation depends on the strength of the omnivoreresource interaction (McLeod et al 2021), but at the whole-network level interactions with other species can stabilize even intrinsically unstable omnivory (Kratina et al 2012). This might be why we find different patterns between a network's motif profile versus a single species motif participation.…”

“…This means that at the network level, omnivory decreased average persistence, while at the species-level, omnivory tended to have positive effect on persistence. Previous studies demonstrate that the network-level effect of omnivory on stability is highly context-dependent (Bascompte et al 2005;Monteiro et al 2016;McLeod et al 2021). The stability of omnivory motifs in isolation depends on the strength of the omnivoreresource interaction (McLeod et al 2021), but at the whole-network level interactions with other species can stabilize even intrinsically unstable omnivory (Kratina et al 2012).…”

Incongruent drivers of network, species and interaction persistence in food webs

“…E7, E8 and E9) is negatively correlated to the persistence of an interaction. Omnivory is an interesting example of conflicting results depending on the level of persistence being examined because the scientific literature lacks a clear consensus about the role of omnivory in communities (Kratina et al 2012, McLeod and Leroux 2020). In particular, there is a body of theoretical research demonstrating that omnivory ranges from being an important stabilizing force within food webs (McCann and Hastings 1997, Neutel et al 2007) to being largely unstable (Levins 1974, Pimm and Lawton 1978, May 2001).…”

“…Theory predicts population dynamics for species within the unique species' positions for isolated motifs (e.g. the resource in an omnivory motif; McCann and Hastings 1997), however, what happens for specific species when these motifs are embedded within larger networks is not as clear (Kondoh 2008, McLeod and Leroux 2020). For example, a species can be a resource in an omnivory motif, but also a resource in a separate exploitative competition motif and the dynamical implications of this dual role which only emerges at the network level are not well understood.…”

Network ecology in dynamic landscapes

“…Directed graphlets could therefore help analyse more complex interactions such as those found in bioinformatics and medicine with comparable ways to align networks [60] and in ecology for more complex trophic interactions (e.g. omnivory [61]).…”