Ecologia hidráulica de árvores amazônicas: da variação intraespecífica ao funcionamento de ecossistemas

Abstract

This thesis helps to understand interactions between environmental gradients and the hydraulic and carbon gain properties of Amazon tree species and how these interactions could affect forest dynamics. Specifically, the first chapter investigated the intraspecific variation of hydraulic and carbon acquisition/conservation traits of two dominant species in a forest in central Amazonia. Examining how environmental and ontogenetic aspects affect the intraspecific variation of these two species. The main finding was a high intraspecific variability of the studied traits. The variation in the hydraulic characteristics of individuals was more associated with soil water variation along topographic gradients. In contrast, the gain/conservation and carbon traits were more associated with ontogenetic and light variations. Furthermore, the existence of these different selection pressures of functional traits of individuals leads to decoupling in the coordination between carbon acquisition/conservation and water use characteristics, breaking the common patterns reported between different species. The second chapter investigated hydraulic functional and carbon gain traits at the community level. Specifically, studied hydraulic and photosynthetic traits related to temperature-sensitive in leaves would be good predictors of diametric growth and, through this, could affect the relative abundance patterns of species. This chapter covered 103 individuals, including 33 species in a forest in central Amazonia. The tested functional traits are good predictors of growth performance, and that the trees that have the lowest growth rates are the rarest in the forest and are also the safest hydraulically. The difference between the ambient temperature and the optimum temperature for photosynthesis for each species affect carbon assimilation, and through this pathway, indirectly affect species dominance. Thus, temperature sensitivity can be used to predict changes in species composition as a function of climate change. The third chapter focused on studing the variation of the tree hydraulics responses in different sites along the Amazon basin, intending to predict the ecosystem's vulnerability to droughts. Specifically, it tested which environmental factors (water table depth, soil fertility, and site precipitation deficit) are most important in defining embolism vulnerability, including 165 trees covering 64 species in 4 sites that combine all variations of the three environmental factors. Shallow water table depth and high soil fertility are more critical in defining a vulnerable hydraulic system than a constant water supply via precipitation (i.e., low seasonal rainfall deficit at the site). The distribution of embolism vulnerability was modeled across the Amazon basin and discussed possible tree strategies to deal with this.