Planting trees to establish an initial canopy is often used to promote tropical forest restoration. Studies on how planted trees impact later forest succession have taken a species-by-species approach. Phylogenetic ecology provides tools to evaluate whether those findings can be extended to closely related species, given the evolutionary conservatism of species interactions. Here we take a phylogenetic ecology approach to study succession in tropical forest restoration. We assessed the species naturally recruiting under a suite of tree species planted in small monoculture plantations to ask 1) if closely related planted tree species will result in similar species composition of understory natural recruits, and 2) what an analysis of the community phylogenetic structure tells us about the forces that shape community assembly.
Results/Conclusions
We found that natural recruitment communities under closely related overstory tree species in the Fabaceae were more similar to each other than expected by chance. It is not clear, however, whether the similarity was driven by broad phylogenetic effects or was specifically an effect of legumes. We predicted that negative biotic interactions should create lower than random coexistence of close relatives, but the phylogenetic distance between most overstory tree species and the species recruiting beneath them was not different from random. However, the phylogenetic structure among species of natural recruits in the understory showed two non-random tendencies; species under legume overstory trees were more distantly related to each other than expected at random, whereas the species recruiting under non-legumes where more closely related. These non-random patterns were likely an effect of the preferential recruitment of the evolutionarily distant Piper clade under legumes, and a potential effect of habitat filtering under non-legumes. Our results show a weak, yet informative phylogenetic signal in the assemblage of communities under trees planted for restoration. This suggests that further work on phylogenetic relationships and ecologically important functional traits may be useful in choosing which species to plant for tropical forest restoration.