Tropical tree species richness and community composition varies along natural and anthropogenic gradients. Microbial pathogens and mutualists, insect herbivores, and predators of insect herbivores may contribute to such variation. Plant-microbial and plant- insect interactions affect the seedling performance and local abundance of tropical tree species. Evidence from temperate systems suggests that variation in the amount of insect herbivory can limit the distribution of plant species. Further, plant-soil feedbacks (PSF), the reciprocal effects between a plant and the soil microbial community it grows in, have been shown to drive the successional trajectories of grassland ecosystems. However, we know little about the role of plant-microbial and plant-insect interactions in determining the dynamics of secondary and mature tropical rainforests. This thesis aims at contributing knowledge to fill this research gap in four data chapters.
Two large greenhouse experiments and two manipulative field studies were performed to address the role of biotic interactions on tropical plant communities in Panama from different perspectives. The greenhouse experiments described in Chapter 2 and 3 tested the variation in PSF with successional stage (i.e. in soils from forests differing in the duration of their recovery since agricultural abandonment), experimental light level, tree species’ association with specific stages of succession, and phylogenetic distance between soil conditioning and succeeding tree species. Chapter 4 assessed variation in seedling vigour of the widespread tropical tree species Lacistema aggregatum across four forest sites along a rainfall gradient within its distributional range and correlated these changes with measurements of pest effects (fungal pathogens and insect herbivores) on plant performance. Chapter 5 explored whether predation pressure on insect herbivores varies with time of the day and host plant pubescence and assessed the effect of model prey shape on estimates of predation in two lowland rainforests.
The results of this thesis are threefold. First, the greenhouse experiments showed that species-specific PSF acted in soils from all successional stages and affected most tree species tested. The occurrence of negative and positive net PSF suggests roles of both microbial pathogens and mutualists in structuring rainforest communities. PSF may contribute to driving the rate and direction of successional tree species turnover by overall more positive PSF occurring at successional stages in which a species is naturally abundant, a lower susceptibility of tree species that are abundant in late successional stages, and a phylogenetic signal in heterospecific PSF. Second, pest effects varied across the rainfall gradient and peaked where plant vigour was highest, providing evidence that pest effects may contribute to determine tree species’ local abundance but no support for pests limiting the distribution of this tree species. Third, this thesis shows that model prey shape may not be as important in prey recognition as assumed in previous studies indicating the need for caution when interpreting model-prey-derived estimates of predation.
The findings of this thesis indicate widespread effects of biotic interactions on plant community dynamics in secondary and mature rainforests. This contributes to closing the knowledge gap on the role of plant-microbial and plant-insect interactions in driving patterns of plant diversity and community structure in tropical rainforests. Its findings contribute to advancing fundamental science on the drivers of tropical biodiversity. Further, they provide information that enhances our ability to predict the consequences of projected changes in climate on tree species distributions and to develop effective restoration strategies to promote the recovery of tropical rainforests after severe disturbance.