2020 ESA Annual Meeting (August 3 - 6)

COS 59 Abstract - Poor soil, rich forest: Leaf nutrients and habitat specialization in an Amazonian forest

Marcel Vaz, Ecology and Evolutionary Biology, UCLA, Los Angeles, CA, José L.C. Camargo, Biological Dynamics of Forest Fragments Project, National Institute for Amazonian Research (INPA) and Smithsonian Tropical Research Institute, Manaus, Brazil, Alberto Vicentini, Instituto Nacional de Pesquisas da Amazônia, Brazil, Alexandre A. Oliveira, Departamento de Ecologia, Universidade de São Paulo, São Paulo, Brazil, Benjamin L. Turner, Smithsonian Tropical Research Institute, Balboa, Panama and Nathan Kraft, Department of Ecology and Evolutionary Biology, University of California, Los Angeles, CA
Background/Question/Methods

Terra firme forests in the Central Amazon are among the most diverse plant communities on Earth, even though they feature extremely weathered and infertile soils. Variation in soil conditions is often driven by local topography, which creates contrasting habitats such as clayey plateaus and sandy valleys. Many have hypothesized that tree species can specialize on these habitats, which could help to explain the diversity of these communities. However, the ecological mechanisms that drive habitat specialization are largely unknown for these forests. Here we suggest that leaf nutrient content may play a fundamental role in restricting the types of soils a tree can grow on. Trees that can produce leaves with low nutrient content might have an advantage on the poor soils of plateau habitats, but because leaf nutrients also limit potential growth, low leaf nutrients should be detrimental in the more fertile soils of valley habitats. To test these predictions, we quantified the leaf content of six macronutrients (N, P, K, Ca, Mg, S) for >1,000 tree species found in a 25-ha permanent plot near Manaus, in Brazil. Finally, we also used spatial analyses to assess habitat specialization for 200 of the most abundant species.

Results/Conclusions

Our multivariate, hierarchical Bayesian model indicates that leaves from plants sampled on the plateau tend to have lower P, K, Ca, and Mg compared to plants sampled in the valley. This trend is in part the result of plastic, within-species variation in leaf nutrients, but interspecific variation was the major source of variation. We also found that legumes behave differently, having higher leaf N and P, but lower Ca and Mg compared to non-legumes, and these differences are more pronounced in the valley. Finally, we found that leaf nutrients are indeed related to habitat association: species with higher average leaf contents of N, P, and K tend to be more restricted to the valleys. These results highlight the role that variation in soil nutrients play in shaping species distribution even on the small spatial scale examined here. The associations found, nonetheless, are not strong enough to completely explain the remarkable alpha diversity of this forest. The leaf nutrient data presented here open a new avenue for exploring the links between soil and leaf nutrients and their effects on plant performance. These findings also contribute to the longlasting debate about the drivers of species diversity and maintenance in tropical forests.