Seed Dispersal near and Far: Patterns across Temperate and Tropical Forests

Abstract

Dispersal affects community dynamics and vegetation response to global change. Understanding these effects requires descriptions of dispersal at local and regional scales and statistical models that permit estimation. Classical models of dispersal describe local or long-distance dispersal, but not both. The lack of statistical methods means that models have rarely been fitted to seed dispersal in closed forests. We present a mixture model of dispersal that assumes a range of disperal patterns, both local and long distance. The bivariate Student's t or ''2Dt'' follows from an assumption that the distance parameter in a Gaussian model varies randomly, thus having a density of its own. We use an inverse approach to ''compete'' our mixture model against classical alternatives, using seed rain databases from temperate broadleaf, temperate mixed-conifer, and tropical floodplain forests. For most species, the 2Dt model fits dispersal data better than do classical models. The superior fit results from the potential for a convex shape near the source tree and a ''fat tail.'' Our parameter estimates have implications for community dynamics at local scales, for vegetation responses to global change at regional scales, and for differences in seed dispersal among biomes. The 2Dt model predicts that less seed travels beyond the immediate crown influence (Ͻ5 m) than is predicted under a Gaussian model, but that more seed travels longer distances (Ͼ30 m). Although Gaussian and exponential models predict slow population spread in the face of environmental change, our dispersal estimates suggest rapid spread. The preponderance of animal-dispersed and rare seed types in tropical forests results in noisier patterns of dispersal than occur in temperate hardwood and conifer stands.

A Spatially Explicit Neutral Model of β-Diversity in Tropical Forests

“…The Gaussian dispersal kernel many theorists employ, however, is often inconsistent with observed seed shadows (Clark et al, 1999). Moreover, different tree species disperse their seeds very different distances, according to very different kernels (Willson, 1993;Clark et al, 1999;Muller-Landau, 2001), even though our neutral model assigns the same dispersal kernel to all trees. Finally, theorists can only solve the neutral model of b-diversity if F ðrÞ expresses an equilibrium between speciation and extinction, an equilibrium that takes long to attain, of the order of 2=n tree generations, in the case of nearest-neighbor dispersal (Section 7, see also Maruyama, 1972).…”

“…Several authors have already explored the implications of limited seed dispersal for the neutral model (Durrett and Levin, 1996;Hubbell, 2001;Chave et al, 2002). The Gaussian dispersal kernel many theorists employ, however, is often inconsistent with observed seed shadows (Clark et al, 1999). Moreover, different tree species disperse their seeds very different distances, according to very different kernels (Willson, 1993;Clark et al, 1999;Muller-Landau, 2001), even though our neutral model assigns the same dispersal kernel to all trees.…”

“…Clark (1998), using the formalism developed by Kot et al (1996), showed that ''fat-tailed'' dispersal kernels, where some seeds disperse quite long distances, can explain the speed with which trees spread in the Pleistocene. One suggested dispersal kernel is (Clark et al, 1999) (1999) fitted dispersal kernels of three types, including (5.1) and the radially symmetric Gaussian, to the distributions of dispersal distances 5100 m of seeds of 24 tree species using a maximum likelihood method. In 14 of these 24 tree species, seed dispersal was best described by dispersal kernel (5.1).…”

“…Ribbens et al (1994) evaluated the effect of differences among tree species in seed dispersal on a model of forest dynamics; Clark et al (1999) fit mathematical formulae to observed dispersal kernels and showed how the dispersal kernel of a tree species might influence its capacity to spread. Several authors have already explored the implications of limited seed dispersal for the neutral model (Durrett and Levin, 1996;Hubbell, 2001;Chave et al, 2002).…”

Spatial patterns of seed dispersal, their determinants and consequences for recruitment

“…For instance, mechanistic models of dispersal (Box 3) can be used to estimate the expected number of seeds each adult would disperse to each trap, and, therefore, the expected sources of seeds in traps 25 . Use inverse methods to fit models specifying distance distributions: this recently developed method relies on numerically intensive calculations of the likelihood of obtaining the observed data on seed-dispersion patterns (Box 1) given a particular model specifying seed shadows of the mapped seed sources 47,50,51 . Some set of plausible phenomenological or mechanistic models (Box 3) having a limited number (typically, two to four) of free parameters must first be specified; the best model and the best parameter values are chosen from among them on the basis of the fit to the data.…”