Interaction Milieu Explains Performance of Species in Simple Food Webs along an Environmental Gradient

Abstract

Species interact with each other in a complex network of relationships that can be modeled as an interaction milieu, i.e. a biotic background with which the species interacts. The interaction milieu is given by the frequency distribution of traits that are important to a given interaction; hence this approach provides a direct link between organism performance, abiotic environmental conditions and interspecific interactions. In this study, we investigate how performance of component species is affected by the interaction milieu along an environmental gradient. We specifically addressed two questions: 1) can the interaction milieu shift species performance away from the niche optimum?, and 2) do species have a higher invasive potential if they have divergent trait values compared to the interaction milieu? We developed a model where a target species showing a given performance response curve to an environmental gradient, i.e., generalist or specialist, interacts with predators and competitors. The predictions of the model were compared with empirical findings on the abundance patterns of species along a salinity gradient in a green beach ecosystem. Green beaches are characterized by their strong gradients and spatial variability in abiotic factors, and their relatively simple food web consisting of Collembola species and their spider predators. First, our results showed that interaction milieu can displace species from their fundamental niche optimum. This led to abundance patterns that cannot be predicted by species performance response curves alone, emphasizing the importance to include ecological interactions. Second, species failed to invade their preferred part of the gradient if predators or competitors shared the target species' preference. Future studies should explore if the concept of interaction milieu can be generalized across trophic levels, and search for traits that are important on deciding the outcome of ecological interactions. Furthermore, the identification of the relevant response traits and their frequency distribution, in combination with trait plasticity of species in interaction milieus might be a step forward to link evolutionary principles to ecological networks, and vice versa.