2018 ESA Annual Meeting (August 5 -- 10)

COS 54-1 - De-coupled hyperdominance in understory and canopy Amazon tree communities

Wednesday, August 8, 2018: 8:00 AM
239, New Orleans Ernest N. Morial Convention Center
Frederick C. Draper, Department of Global Ecology, Carnegie Institution for Science; International Center for Tropical Botany, Florida International University, Christopher Baraloto, INRA, UMR EcoFOG, Kourou, FL, French Guiana, Gregory P. Asner, Department of Global Ecology, Carnegie Institution for Science, Stanford, CA, Flávia R. C. Costa, Coordenação de Pesquisas em Ecologia, Instituto Nacional de Pesquisas da Amazônia (INPA), Manaus, Brazil, Oliver Phillips, University of Leeds, Gabriel Arellano, Tropical Research Institute, Smithsonian Institution, Washington, Manuel J. Macia, Departamento de Biología, Universidad Autónoma de Madrid, Madrid, Spain, Alvaro J. Duque, Universidad Nacional de Colombia, Medellin, Colombia and Juliana Scheitti, Coordenação de Biodiversidade, Instituto Nacional de Pesquisas da Amazônia
Background/Question/Methods

Amazonian forests are the most species rich ecosystems on earth, yet relatively little is known of how this diversity is structured at a basin-wide scale. Recent efforts using extensive plot networks have demonstrated that despite the huge diversity of Amazon tree communities, just 227 so-called “hyperdominant” species account for 50% of all individual trees. However, these studies have excluded trees smaller than 10cm in diameter, thereby excluding some of the most widespread and diverse Amazonian tree lineages. Here we use a new plot network “Rede-Gentry”, to quantify basin-wide patterns of dominance in sub-canopy tree communities for the first time. Rede-Gentry is a collaborative, basin-wide network of over 1000 forest plots (mainly 0.1 ha in size) in which all stems over 2.5 cm in diameter have been measured and identified. Using this network, we quantified the abundance of over 4000 Amazon tree species, from which we identified species that were hyperdominant, accounting for half of all stems in the sub canopy. We tested how these sub-canopy hyperdominants contrasted with hyperdominant canopy trees, in taxonomic, functional and phylogenetic identity.

Results/Conclusions

We demonstrate that, like the canopy, the sub-canopy is dominated by a small fraction of hyperdominant species which together account for 50% of stems. However, sub-canopy hyperdominants are phylogenetically distinct from previously identified canopy hyperdominants, with families such as Myrtaceae, Melastomataceae and Piperaceae comprising multiple hyperdominants only in the sub-canopy. We found that most sub-canopy hyperdominant species were only dominant in a single region and habitat type. Our results demonstrate that hyperdominance is pervasive across the Amazon tree flora in both canopy and sub-canopy trees. However, the identity of hyperdominant species is largely decoupled across size classes. These findings highlight the arbitrary nature of a 10 cm diameter cut-off; by limiting the scope of large scale floristic analyses to only stems greater than 10 cm, previous studies have overlooked some of the most dominant Amazonian species that represent original evolutionary lineages and functional strategies.