Heterospecifics matter: Density dependent survival of seedlings in a tropical forest

Background/Question/Methods

Understanding species coexistence requires identifying the effects on individual performance of size-(a)symmetric and density dependent interactions with conspecific and heterospecific neighbors. Knowledge on such interactions in species-rich tropical forests is biased. Heterospecific effects are particularly underestimated because they vary with ecological similarity between interacting species, which is usually poorly characterized. This is unfortunate because coexistence is the net outcome of multiple mechanisms involving heterospecifics. To address these issues, we analyzed 15 years of survival of 36,225 first-year seedlings of 222 species in the tropical forest of Barro Colorado Island, Panama. We analyzed the effects of the density of conspecific and heterospecific seedling and larger neighbors, and of the phylogenetic relatedness to heterospecific neighbors. We used standard summary metrics of phylogenetic relatedness calculated over all neighbors but, unlike similar studies, also separate heterospecifics into groups that correspond to the divergence time of major taxonomic ranks. This permits us to evaluate effects of phylogenetic relatedness at scales that differ in their link to ecological similarity and in the likelihood of species interactions to occur.  The link between phylogenetic relatedness and functional similarity was also tested by analyzing phylogenetic conservatism of nineteen traits related to resource capture, defense and stress tolerance.

Results/Conclusions

Test of trait conservatism confirmed the use of phylogenetic relatedness to approximate functional similarity. Models using standard summary metrics of phylogenetic relatedness were outperformed by the approach that divided heterospecifics into groups of phylogenetic relatedness. Using such alternative approach, we confirmed strong conspecifics effects, mainly from seedling neighbors.  Because competition is size-asymmetric in closed canopy forests, effects of large neighbors should be stronger than effects of seedlings neighbors. The stronger effects of conspecific seedlings relative to large conspecific neighbors suggests that host-specific pests shared among seedlings rather than resource competition underlies this result. Unlike other studies, large heterospecific neighbors had negative density dependent effects that were comparable to the effects of large conspecific neighbors. This suggests that resource limitation rather than host-specific enemies explain effects of large neighbors on seedling survival.  Survival nonetheless increased near neighbors that were closely related to the focal individual. This suggests that shared habitat associations and/or mutualists have stronger effects on seedling survival than shared pests or resource requirements, which would cause survival to decrease . Improved performance near similar neighbors is an emerging pattern in the literature. Our study highlights that complex but explainable neighborhood interactions determine individual performance in species rich forests.