Integration of ecosystem engineering and trophic effects of herbivores

Herbivores affect vegetation in a variety of ways. involving both trophic a nd ecosystem engineering interactions, but the study of these different int eraction types has rarely been integrated. The aim of this study was to inv estigate both the trophic and engineering effects of herbivores on plant co mmunities in the Negev desert, Israel, and to promote an integrative approa ch to the study of herbivore effects in ecosystems. First, we summarise pre vious studies of the Indian crested porcupine (Hystrix indica), which show that in digging for food, porcupines excavate soil pits, which accumulate r esources and seeds resulting in marked changes in plant species richness, d ensity and biomass. By contrast, their trophic effect, via consumption of b ulbs, has little impact on populations of perennial plants. Second, we pres ent an empirical study of the trophic and ecosystem engineering effects of harvester ants (Messor spp.). An exclusion experiment, using barriers to re strict ant access, failed to reveal any significant effect of seed collecti on by harvester ants on plant species incidence (proportional occurrence in samples) or abundance (number of individuals). However, we show that veget ation on nest mounds of M. ebeninus differs in plant density, species richn ess and biomass from that on undisturbed soil. An analysis of incidence and abundance responses of individual plant species suggests that the observed differences in vegetation resulted from multiple interacting mechanisms. T he case studies highlight that many interactions between herbivores and pla nt communities can occur simultaneously, and that ecosystem engineering and trophic processes can be closely associated. resulting from single actions of herbivores. We propose a conceptual framework that classifies the range of possible trophic and engineering interactions between herbivores and pl ant communities with respect to the level of association between trophic an d engineering effects. The framework is presented as an aid to the design a nd interpretation of studies of interactions between herbivores and plant c ommunities, and promotes integrative research into the roles of herbivores in ecosystems.