Premium
Macroecology of seed banks: The role of biogeography, environmental stochasticity and sampling
Author(s) -
Jabot Franck,
Pottier Julien
Publication year - 2017
Publication title -
global ecology and biogeography
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.164
H-Index - 152
eISSN - 1466-8238
pISSN - 1466-822X
DOI - 10.1111/geb.12631
Subject(s) - species richness , macroecology , ecology , biodiversity , metacommunity , vegetation (pathology) , beta diversity , geography , biogeography , soil seed bank , species diversity , sampling (signal processing) , biology , biological dispersal , medicine , population , demography , seedling , filter (signal processing) , pathology , sociology , computer science , horticulture , computer vision
Aim The study of seed banks has been mainly conducted at local scales, thereby hampering our general understanding of the assembly processes of these biodiversity reservoirs. Here, we aim to document worldwide macroecological patterns of seed bank diversity and of their similarity with aboveground vegetation. Our second aim is to investigate the likely drivers of these macroecological patterns and to lay the foundation of a metacommunity theory of seed banks. Location Worldwide. Time period 1989–2005. Major taxa studied Plants. Methods We compiled a worldwide dataset of 130 seed banks located in grasslands. We assessed the likely drivers of seed bank diversity and similarity with aboveground vegetation, using structural equation modelling. We then developed a time‐averaged neutral model of coupled seed bank–vegetation (S‐V) dynamics that includes the effect of environmental stochasticity, and we compared its predictions with empirical findings. Results We found evidence for two weak latitudinal gradients in seed bank diversity and S‐V similarity, with larger species richness and smaller similarity closer to the tropics. We then showed that seed bank richness and S‐V similarity are correlated with the following four distinct drivers: local environmental variability, plant regional diversity, seed bank density and sampling area. Finally, we showed that the predictions of the time‐averaged neutral model are remarkably in accord with empirical observations. Main conclusions Our results lay the foundations of a metacommunity theory for seed banks. They challenge the standard view that seed banks are principally structured by environmental variability, by highlighting the additional key roles of dispersal and stochastic sampling on seed bank diversity patterns.