Conspecific and heterospecific crowding facilitate tree survival in a tropical karst seasonal rainforest

Tree mortality plays a vital role in forest dynamics and contributes to species coexistence and community assembly. However, the mechanisms that control tree mortality are poorly understood, particularly in species-rich tropical forests. Size-dependent abiotic constraints and biotic interactions act simultaneously to cause tree mortality in natural forests. However, these drivers are often studied independently, which can limit our understanding of how they interact to affect tree mortality in natural forests. We employed a hierarchical Bayesian logistic regression modeling approach to quantify the simultaneous effects of tree size, topography, neighborhoods, and their interactions on tree mortality in a 15-ha fully mapped tropical karst seasonal rainforest in Southern China. Of the variables tested, tree size exhibited the most consistent and strongest negative effect on tree mortality, while topography and neighborhood competition were of secondary importance. Neighbors, particularly heterospecific neighbors, had a facilitative effect on lowering mortality, while only several species showed conspecific negative density dependent on mortality. The topographical wetness index had a significant positive effect on tree mortality. Size-dependent and neighborhood competition-induced tree mortality was stronger for shade-intolerant than for shade-tolerant species, and the size-dependent mortality was the strongest for shrub species among the three growth forms. Our results indicated that the size-dependent effect was the dominant cause of tree mortality; however, neighbors, particularly heterospecific neighbors, had a facilitative effect on lowering tree mortality. Furthermore, the relative importance of these mechanisms for tree mortality differed within life-history strategies guilds. Our study highlighted the value of simultaneously considering the individual and interactive effects of multiple mechanisms toward understanding the dynamics and coexistence of highly diverse metacommunities in harsh environments.