AUTHOR=Oliveras Imma , Bentley Lisa , Fyllas Nikolaos M. , Gvozdevaite Agne , Shenkin Alexander Frederick , Peprah Theresa , Morandi Paulo , Peixoto Karine Silva , Boakye Mickey , Adu-Bredu Stephen , Schwantes Marimon Beatriz , Marimon Junior Ben Hur , Salinas Norma , Martin Roberta , Asner Gregory , Díaz Sandra , Enquist Brian J. , Malhi Yadvinder 

TITLE=The Influence of Taxonomy and Environment on Leaf Trait Variation Along Tropical Abiotic Gradients

JOURNAL=Frontiers in Forests and Global Change

VOLUME=Volume 3 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/forests-and-global-change/articles/10.3389/ffgc.2020.00018

DOI=10.3389/ffgc.2020.00018

ISSN=2624-893X

ABSTRACT=Abstract
Deconstructing functional trait variation and co-variation across a wide range of environmental conditions should increase the mechanistic understanding of community assembly processes and improve current parameterization of dynamic vegetation models. Here, we present a study that deconstructs leaf trait variation and co-variation to iithin-species, taxonomic-interspecific, and plot-environment components comparing three tropical environmental gradients  in Peru, Brazil and Ghana. We measured photosynthetic, chemical and structural leaf traits using a standardized sampling protocol, totalling more than 1,000 individuals belonging to 367 species sampled. Variation associated with the whole interspecific taxonomic component (species+genus+family) for most traits was relatively consistent across environmental gradients, but intra-specificwithin-species variation and the plot-environment variation was strongly dependent on the environmental gradient. Trait-trait co-variation was also strongly linked to the environmental gradient where the traits were measured, although some traits had consistent co-variation components irrespective of environmental gradient. Our results demonstrate that filtering along gradients is mostly expressed through trait intra- and interspecifictaxonomic variation, but that trait co-variation is strongly dependent on the local environment, and thus global trait co-variation relationships might not always apply at smaller scales.