The effects of forest fragmentation on seed dispersal and neotropical forest regeneration are poorly understood. Research in the Amazon has shown that potential seed dispersers (primates, understory frugivorous birds, and bats) have lower densities in forest fragments compared to continuous forest. In contrast, potential seed predators (terrestrial small mammals) are more diverse and abundant in fragments. These studies suggest that seed dispersal may be lower, and seed predation higher in small fragments, however no studies have tested this. The proposed project will be the first to assess the effect of forest fragmentation on tree regeneration dynamics in the central Amazon and quantitatively relate these effects to modifications in plant-animal interactions that are hypothesized to influence tree regeneration. We will quantify tree seedling species richness and composition in forest fragments (1, 10, and 100 ha) and continuous forest. We will simultaneously measure seed rain, seed predation, and herbivory. We will test the hypothesis that forest fragmentation negatively affects forest regeneration through losses in frugivores and simultaneously test the alternate hypotheses that seed predation, herbivory, or edge effects negatively impact upon biodiversity and composition of tree regeneration.
The discovery and measurement of dispersal limitation of forest regeneration will have major implications for land management in the neotropics, as it will contribute to estimates of the minimum fragment size and the maximum fragment isolation that can allow sustained regeneration. The research will also identify links between seed dispersers, predators, and fragment size that can be used in predictive models to determine the effects of land use practices on native forest species.
