98th ESA Annual Meeting (August 4 -- 9, 2013)

COS 55-10 - Quantifying the roles of shade tolerance and successional niches in tree species coexistence

Wednesday, August 7, 2013: 11:10 AM
L100C, Minneapolis Convention Center
Tao Zhang and Jeremy W. Lichstein, Department of Biology, University of Florida, Gainesville, FL
Background/Question/Methods

Different hypotheses have been developed to explain biodiversity in forests. A traditional view is that a tradeoff between low-light survivorship (LLS) and high-light growth (HLG), which has been observed in some forest communities, allows species with different successional niches to coexist. However, some well-studied forest communities provide little empirical support for the existence of this tradeoff, which has led to the hypothesis that high-dimensional tradeoffs are required for many species to coexist. Here, we combine theory derived from the Perfect Plasticity Approximation (PPA) forest dynamics model, forest inventory data, and a widely used shade-tolerance classification (Silvics of North America) to explore the role of LLS in determining late-successional competitive ability, and the role of successional niches in tree species coexistence. We used a metacommunity version of the PPA model (in which stand-replacing disturbances occur randomly in time and space) to quantify the number of successional types that can coexist. We used a numerical algorithm to solve for the multi-species transient dynamics that a single patch would undergo when the entire metacommunity is at equilibrium.

Results/Conclusions

A key result of PPA theory is that the late-successional dominant species is the species that, if grown in monoculture, would have the highest equilibrium height of canopy closure (“Z*”: the height of the shortest tree in the upper canopy, receiving full sunlight from above). This index of late-successional competitive ability is a function of multiple demographic and allometric parameters, in addition to LLS. Therefore, the absence of a HLG-LLS tradeoff does not necessarily imply that successional niches are unimportant in forest community structure, because late-successional dominance does not require high LLS. Empirically, we show that LLS and Z*hat are positively correlated within most, but not all, ecoprovinces in the eastern U.S. Both indexes are strongly correlated with the Silvics of North America shade-tolerance classification. Solutions to the PPA metacommunity model showed that, using growth and mortality rates typical of temperate forests and realistic disturbance rates, only 3 to 4 successional types can stably coexist due to successional niche differentiation alone. Therefore, although successional niches may be a widespread and important factor in forest community assembly, these niches alone typically allow only a modest number of tree species to coexist. Coexistence of larger numbers of tree species likely depends on other mechanisms.