Hierarchical mechanisms of spatially contagious seed dispersal in complex seed‐disperser networks

Abstract

Abstract. Intra-and interspecific spatially contagious seed dispersal has far-reaching implications for plant recruitment, distribution, and community assemblage. However, logistical and analytical limitations have curtailed our understanding concerning the mechanisms and resulting spatial patterns of contagious seed dispersal in most systems and, especially, in complex seed-disperser networks. We investigated mechanisms of seed aggregation using techniques of spatial point pattern analysis and extensive data sets on mutispecific endozoochorous seed rain generated by five frugivorous mammals in three Mediterranean shrublands over two seasons. Our novel analytical approach revealed three hierarchical and complementary mechanisms of seed aggregation acting at different levels (fecal samples, seeds, pairs of seed species) and spatial scales. First, the three local guilds of frugivores tended to deliver their feces highly aggregated at small and intermediate spatial scales, and the overall pattern of fecal delivery could be described well by a nested doublecluster Thomas process. Second, once the strong observed fecal aggregation was accounted for, the distribution of mammal feces containing seeds was clustered within the pattern of all feces (i.e., with and without seeds), and the density of fecal samples containing seeds was higher than expected around other feces containing seeds in two out of the three studied seeddisperser networks. Finally, at a finer level, mark correlation analyses revealed that for some plant species pairs, the number of dispersed seeds was positively associated either at small or large spatial scales. Despite the relatively invariant patterning of nested double-clustering, some attributes of endozoochorous seed rain (e.g., intensity, scales of aggregation) were variable among study sites due to changes in the ecological context in which seeds and their dispersers interact. Our investigation disentangles for the first time the hierarchy of synergic mechanisms of spatially contagious seed dispersal at a range of spatial scales in complex seeddisperser networks, thus providing a robust and widely applicable framework for future studies.

Genetic rescue by distant trees mitigates qualitative pollen limitation imposed by fine‐scale spatial genetic structure

“…Specifically, trees exhibit aggregation at spatial scales of approximately 25 m in Matasgordas (6–8 reproductive individuals within approximately 25 m; Fedriani et al, ) and 18 m in Reserva (10–11 reproductive individuals; A. R. Castilla, unpublished data). The establishment and persistence of such tree clusters arise from dispersal limitation, spatial aggregation of dispersed seeds, and clonal growth (Castilla, Godoy, Delibes, Rodriguez‐Prieto, & Fedriani, ; Fedriani & Wiegand, ; Fedriani et al, ). The density of reproductive individuals is greater in the Matasgordas population than in the Reserva (3.096 ± 0.396 and 0.089 ± 0.054 individuals/ha; Garrote et al, ).…”

Assisting seed dispersers to restore oldfields: An individual‐based model of the interactions among badgers, foxes and Iberian pear trees

“…25 m; Fedriani et al., ). The establishment and persistence of such Iberian pear clusters arise from dispersal limitation and spatial contagion of dispersed seeds (Fedriani & Wiegand, ; Fedriani et al., ). The Doñana population of Iberian pear has limited reproduction and regeneration ability (Fedriani et al., , ) and shows a marked left‐skewed demographic structure, with many individuals in older age classes, few juveniles, and even fewer seedlings and saplings (Authors, unpubl.…”

Unravelling conflicting density‐ and distance‐dependent effects on plant reproduction using a spatially explicit approach

“…( r ), I mm ( r ) and C m , K ( r ), being the fifth lowest and highest values of the summary statistics among the 199 simulations of the null model (e.g. Wiegand, He & Hubbell ; Fedriani & Wiegand ). We observe a departure from the null model at particular distances r if the summary statistic of the observed data is outside the simulation envelopes.…”

“…; Raventós, Wiegand & De Luis ; Getzin et al . ; Wiegand, He & Hubbell ; Fedriani & Wiegand ). This allows us to find out, for example, whether pollination success is larger for trees that have more neighbours compared to trees with fewer neighbours.…”

“…Analysis of such quantitatively marked patterns allows detection of correlations in PRS across a range of spatial scales (e.g. Illian et al 2008;Ravent os, Wiegand & De Luis 2010;Getzin et al 2011;Wiegand, He & Hubbell 2013;Fedriani & Wiegand 2014). This allows us to find out, for example, whether pollination success is larger for trees that have more neighbours compared to trees with fewer neighbours.…”