The role of landscape and species attributes in insect community assembly, population genetics and plant-insect interactions in expanding quercus ilex forests

Resumen

The mechanisms driving species colonization and coexistence in habitats, is one of the main puzzles in community ecology. Several theories, including niche, neutral and coexistence theories, have tried to explain the biodiversity observed in natural ecosystems as the result of deterministic, neutral, or a combination of both processes, respectively. This biodiversity will assemble in communities and ultimately establish biotic interactions, which will play an important role in the ecosystem dynamics and its resilience. Species interactions can only be fully approached by the double perspective of landscape ecology and metapopulation/metacommunity dynamics, and therefore both, landscape attributes and species’ traits must be taken into account. Loss of communities and biotic interactions has been thoroughly studied in the context of shrinking and fragmented habitats, as it is one of the main threats to global biodiversity; however, the assembly of communities and biotic interactions in new expanding natural ecosystems is a much less explored topic. In the current context of global change, land transformation has resulted in loss and fragmentation of habitat in many different ecosystems, however, recovery of natural habitats has also been observed. Particularly in Europe, cropland abandonment has led to an intense process of forest expansion and nowadays we can find patchy afforested areas all over the continent in which an ongoing process of assembly is taking place. Especially in these patchy habitats, neutral processes modulated by spatiotemporal landscape attributes as well as dispersal ability of organisms could be playing a key role in the assembly of communities and biotic interactions. Nevertheless, deterministic processes might as well be shaping communities and interactions at other spatial scales The aim of this thesis has been, therefore, to deepen in the knowledge of factors driving community and interaction assembly, taking as study system the plant-insect interaction established between species that have successfully colonized these expanding forest patches, specifically Quercus ilex, a key species of the Mediterranean forests, and its community of herbivore (folivore and granivore) insects. As a first objective, this thesis proves the impact of spatiotemporal attributes at the tree scale on the assembly of the Lepidoptera community. Chapter 2 shows a complex interplay of age, canopy size and connectivity of trees interspersed in an expanding savannah-like Q. ilex landscape, leading to more abundant, richer and more diverse Lepidoptera communities in older, larger and more connected trees, while alpha and beta diversity depending exclusively on the distance between canopies, therefore validating the importance of neutral and random processes on the assembly of the herbivore community. As a second objective, this thesis explores the impact of spatiotemporal and genetic attributes at the tree scale on herbivory. Chapter 3 shows a direct effect of tree genetics on the leaf damage suffered by holm oaks, as well as an edge effect benefiting insects at the core of the patch, while an indirect effect of age and connectivity, resulting in associational resistance, is revealed among mature developmental stages of holm oak. Additional support of tree characteristic impact on herbivory is found in Chapter 4, where seed-predation is regulated by acorn size and tree connectivity, while it does not change between habitats or landscapes at the local and regional scales, due to compensatory dynamics of the different species within the seed-predator guild. This homogeneity in the interaction, contrasts with the heterogeneity shown by seed-predator assemblages at the three spatial scales, with colonization credits of the poorer dispersers in the most new and isolated patches. Therefore, results of Chapter 4 evidence, indeed, the importance of studying community and interaction assembly at different spatial scales as well as revealing different seed-predator assemblages at the local scale but zero-sum landscape effects on seed-predation at the landscape scale. In addition to effects on the composition of the community, the colonization process mediated by landscape and species’ attributes can also have genetic effects. Population genetic theory predicts a strong correlation between population size and genetic variation, which in turn, is expected to correlate with fitness and adaptation potential. In Chapter 5 we explore the effects of age and connectivity of the habitat in combination with the dispersal vs. dormancy trade-off exhibited by some seed-predator species. Results reveal significant genetic structure and low genetic variability only for the poor disperser Curculio elephas, as well as restricted gene flow between old and new-isolated patches, in consonance with colonization credits found for this species in Chapter 4. In spite of the founder effects, consistent larger populations of C. elephas compared to the better disperser C. glandium suggest that other factors are favoring the fitness of the former species Overall, results included in this thesis provide a detailed insight on the processes that, at different spatial scales, are mediating the Quercus ilex – herbivore assembly in expanding forests, revealing the participation of all, niche, neutral and random processes. Findings also suggest that the patchy arrangement of these expanding habitats are favoring the existence of different herbivore assemblages, but with a resultant zero-sum impact of herbivory on holm oaks. Differential colonization dynamics, mediated by differences in species dispersal ability, have already left a genetic signature, which may impact seed-predator fitness and potential of adaptation. The information provided in this thesis may be of great help for the management of both, Q. ilex and herbivore populations in expanding forests, supporting decisions such as size/shape patch maintenance of enhancement/reduction of inter-patch connectivity.