Above- and belowground herbivores influence each other mediated by the host plant. Most of the studies on this matter have been conducted in controlled and homogeneous laboratory conditions. However, the reality is that species interact within dynamic, heterogeneous and complex landscapes. Moreover, the dispersal capacity of species is obviously different. While plants and aboveground insect herbivores have the chance to move over long distances by means of flight or seed dispersal, belowground herbivores are limited in their movements. This situation results in an asymmetrical shifting mosaic in time of species interactions, where organisms are present in some patches and not in other. Nevertheless, it has never been addressed how species interactions, such as those of below- and aboveground herbivores, are affected by their own mobility within structurally complex landscapes and what are the evolutionary consequences for herbivory-related life history traits (e.g. plant toxin production and development of resistance against toxins by herbivores). We developed an individual based simulation model in which we addressed the effect of landscape structure on the population dynamics and the evolution of toxin production/resistance of a plant and two herbivores, an aboveground and a belowground. Simulations were run under different landscapes testing different mobility for the species, with belowground herbivores always having a very restricted movement and, conversely, a continuum of dispersal ranges for plant seeds and aboveground herbivores. The interaction between the plant and both herbivores was studied under a trade-off scenario where the multiplication of each of the three species was modified in time as a function of the development of toxins or resistance.

Results/Conclusions

The outcome of the model indicated that the final population size of each species, the level of toxin secreted by the plant, and of resistance to plant toxins of the herbivores were strongly influenced not only by the landscape structure, but ultimately by the dispersal capacities of the aboveground herbivore. The higher is the landscape availability and connectivity, the higher are the toxin levels of the plant, due to the fact that the chance for the three organisms to interact is very high. In patchy landscapes (low spatial connectivity and habitat availability) the most important parameter determining the outcome of the interactions for the three species is the dispersal of the aboveground herbivore, but interestingly, not always in a linear way. Our model consequently provides evidence of the importance of landscape structure and dispersal for the evolution of plant-herbivore interactions.