Tree size distribution, population trend and shade tolerance

Background/Question/Methods Tree size distributions have long been of interest to ecologists and foresters because they reflect fundamental demographic processes. Previous studies have shown that size distributions are often associated with population trends and/or with the degree of shade tolerance. In this study, we tested these associations using data from two consecutive censuses of a 20 ha plot in southern China. We focused on the 31 species with over 500 individuals, species that varied widely in growth form and shade-tolerance. We classified the species into modal and reversed J shaped by comparing the whether the smallest size class has significantly more individuals than the other size classes and tested whether size distribution was associated with shade tolerance and population trend with chisquare testes. We further tested whether modal species and reversed J species had significantly different recruitment rate, mortality rate and per capita population growth rate by Wilconxon rank sum tests. We made projection on the population using matrix model and tested whether the fate of the population relied on the size distribution. We fit size-dependent growth and mortality function with LOESS method and found whether they had strong links with size distribution and obtained the equilibrium size distribution according to the one – dimensional continuity equation of fluid dynamics and tested whether the modal species and reversed J species were equally at equilibrium. 

 Results/Conclusions Eight species had unimodal size distributions, while 23 had reversed J shaped distributions. The type of size distribution was not associated with per capita population growth rates, or with shade-tolerance. Modal species, as a whole, had lower recruitment and mortality rates than reversed J species. All modal species had U-shaped size-dependent mortality and growth functions, with minimum values of both mortality and growth at intermediate size classes. The matrix population model predicted that the two types of species wouldn't have significant different population growth rate until the end of the century. The prediction of size distribution from growth and mortality functions worked better for reversed J species than modal species, indicating that modal species were less at equilibrium than reversed J species. These results show that modal size distribution do not necessarily indicate either population decline or shade-intolerance. Instead, these species in our study were characterized by a life history strategy of extreme conservatism in an intermediate size class, leading to very low growth and mortality in that size class, and thus to a peak in the size distribution at intermediate sizes.