Características de dispersão de sementes e frutos ao longo dos biomas brasileiros: explorando tendências, predizendo e mapeando correlatos ecológicos

Abstract

Seeds and fruit traits are essential for numerous functions along the plant regeneration process. Despite their remarkable importance, such traits have been poorly understood concerning leaf and root traits. For instance, it is still unknown how dispersal seed and fruit traits covary within different environments and dispersal strategies. Furthermore, there is a lack of congruence regarding the key ecological factors shaping the distribution of seed dispersal traits. Thus, I address these knowledge gaps by gathering seed and fruit traits and environmental variables from 301 old-growth vegetation plots across the four largest Brazilian biomes (Amazon, Atlantic Forest, Cerrado, and Caatinga). In Chapter 1, I explored relationships for a set of six morphological traits (seed mass, seed and fruit width and length, and seed number) at the species level, differentiating them among biomes and/or dispersal modes. In the second chapter, I examined the distribution of dispersal modes and seed mass across Brazilian biomes and identified the main climatic, edaphic, and dispersal agent factors that predict the dominance of the most common dispersal modes and seed mass. In the first chapter, I found that the morphological seed and fruit spectrum is divided into two principal components that represent the size traits (seed mass and seed and fruit dimensions) and the seed number per reproductive unit. These two components of variation are maintained regardless of biomes and dispersal modes. However, I observed that seed mass from wind-dispersed species varied independently of size traits within biomes (except in the Amazon). In the second chapter, I found that climate, in terms of its variation and precipitation levels, was the primary factor in predicting the dominance of seed mass and dispersal modes, followed by soil properties or dispersal agents. Regarding the ecological hypotheses, the resource-availability hypothesis was not well- supported in predicting the distribution of dispersal modes. However, the distribution of wind-dispersed species was predicted by the mean annual wind speed (disperser- availability hypothesis), while heavy-seeded species increased under conditions typically found in shade environments (recruitment-hypothesis) and poor soils in terms of bulk density. New advances are needed to gain a deeper understanding of the mechanisms underlying the complex relationships between frugivores and the dominance of animal-dispersed species. Finally, I discuss the implications of my results in helping anticipate how future climate changes and anthropogenic disturbances will impact ecosystem functioning.