Shift of spatial patterns during early recruitment in Fagus sylvatica: Evidence from seed dispersal estimates based on genotypic data

A major challenge in population ecology of long-lived plants is to disentangle the intricate effects of seed dispersal and seedlings mortality on spatial and genetic structure of recruited seedlings. A common way to address this issue is to compare dispersal patterns among seeds (i.e. basic dispersal) and established seedlings (i.e. effective dispersal). High density-dependent mortality near adults is expected to shift the mode of the effective dispersal distance away from the mother-plant relative to the mode of the basic dispersal distance. Here, we proposed a new approach to address this issue by using seedlings cohorts of different age and by combining parentage analyses to retrieve effective seed dispersal and Spatially Explicit Mating Models (SEMM) to estimate basic seed dispersal. We applied this approach in two seedling cohorts at each of two plots of Fagus sylvatica L., relying on adult and seedling genotypes at 10 microsatellite loci. Basic seed dispersal was primarily local, with mean distances of 12-25 m (young cohort) and 22-25 m (old cohort). The effective seed dispersal distance increased between young and old cohorts from 33 m to 48 m at site I and from 23 m to 33 m at site 2. The hypothesis of density-dependent mortality was supported at site 2 but not at site 1 possibly because of different ecological conditions among sites. Our results shed light on the interplay between seed dispersal and density-dependent mortality in F. sylvatica. Moreover, the new approach proposed here alleviates general problems associated to synchronic studies comparing spatial patterns across different-age cohorts at a single point in time. (C) 2013 Elsevier B.V. All rights reserved.