Science Rendue Possible
Pérez-Hernández, C. X., W. Dáttilo, A. M. Corona-López, V. H. Toledo-Hernández, and E. del-Val. 2022. Buprestid trophic guilds differ in their structural role shaping ecological networks with their host plants. Arthropod-Plant Interactions. https://doi.org/10.1007/s11829-022-09933-w
Plant–herbivore relationships involve a significant amount of global biodiversity within complex interaction networks. Buprestidae (Coleoptera) are highly specialized herbivores, and several species have important economic and ecological impacts. We used tools derived from network theory to evaluate the structure of a plant-buprestid metaweb at three different organizational levels (network, subnetwork, and species-levels) and test whether trophic guilds and taxa differ in their patterns within the network. We also tested whether taxonomically closely related buprestid species are more likely to share the same host plant species. We found that the plant-buprestid metaweb exhibits a non-nested and significantly highly modular pattern, and most buprestid and host plant species have specialized interactions. Florivorous buprestids showed the highest diversity of host preferences and, together with Fabaceae, were the most important for the network structure as they are highly connected species. Leaf-mining buprestids had the most extreme interaction pattern among trophic guilds, with high modularity and specialized interactions. We also found a low probability to share the same group of host plants among buprestids, which decreased with taxonomic distance. Our findings uncover patterns within a plant–herbivore network at large spatial scales and at different taxonomic levels, which are shaped by the diversity of host and resources preferences, more than taxonomic relatedness. Those network patterns might reflect different ecological roles for each trophic guild and taxa. We highlight the relevance of considering the diversity of feeding habits within networks of a single type of interaction and emphasize the importance of analyze network patterns at high levels of organization.