The functional form of specialized predation dramatically affect whether Janzen-Connell effects can prevent competitive exclusion

Daniel J. Smith, Committee on Evolutionary Biology, University of Chicago, Chicago, IL and J. Timothy Wootton, Department of Ecology and Evolution, The University of Chicago, Chicago, IL

Background/Question/Methods
The Janzen Connell Hypothesis (JCH) is a long-studied species coexistence mechanism frequently invoked to explain the high diversity of tropical forests. The conditions necessary for the JCH to operate, the presence of distant-dependent specialized predation, enjoy strong empirical support and theoretical work demonstrates that Janzen-Connell Effects (JCEs) are effective in preventing extinction due to ecological drift. However, recent models that consider inter-specific variation in JCE susceptibility or inter-specific variation in intrinsic fitness indicate that JCEs may be unable to reduce competitive differences to the point of community-wide coexistence. A potential weakness of recent JCH-type models analyzing the deterministic ability of JCEs to promote diversity is the chosen functional form of specialized predation pressure. These papers assume that conspecific seeds and seedlings directly beneath a parent tree or within a fixed neighborhood of the parent tree are less likely to survive by a fixed proportion. In this functional form of specialized predation, JCEs act in a binary manner: they occur in full effect on patches if one or more conspecific adults are within the defined neighborhood surrounding the patch, and do not occur if adults are not within the neighborhood. I refer to this model as the "fixed neighborhood model". However, empirical evidence indicates that the strength of predation pressure that seeds and seedlings experience in a particular location increases with the number of conspecific trees in that local area scaled by distance. I refer to this as the ``additive predation model''. I use deterministic Ordinary Differential Equation (ODE) approximations of spatially explicit JCH-type models and compare the fixed neighborhood model to the additive predation model. These approximations accurately compare to extent to which each JCE functional form promotes deterministic coexistence

Results/Conclusions
I demonstrate that the additive predation model promotes considerably higher than the fixed neighborhood model when species vary in either intrinsic fitness of JCE susceptibility. While these results do not necessarily demonstrate JCEs to be of singular importance in the maintenance of tropical rainforest diversity, they do demonstrate that recent results indicating that JCEs weakly promote coexistence may be the consequence of modeling assumptions. This study highlights the need to determine the exact functional form of distance dependent predation in order to elucidate its role in facilitating species coexistence.