COS 20-5 - Why do leaf venation networks have loops? Testing hypotheses with an Andes-Amazon elevation gradient

Monday, August 7, 2017: 2:50 PM
D138, Oregon Convention Center
Benjamin W. Blonder, UA Science: Sky School, University of Arizona, Tucson, AZ, Norma Salinas, Sección Química, Universidad San Antonio Abad del Cusco, Lima, Peru, Lisa Patrick Bentley, Sonoma State University, CA, Alexander Shenkin, University of Oxford, Oxford, United Kingdom, Percy Chambi Porroa, Universidad Nacional de San Antonio Abad del Cusco, Peru, Yolvi Valdez Tejeira, Universidad Nacional de San Antonio Abad del Cusco, Tatiana Boza Espinoza, University of Zurich, Switzerland, Gregory R. Goldsmith, Ecosystem Fluxes Group, Paul Scherrer Institute, Villigen, Switzerland, Lucas Enrico, University of Cordoba, Argentina, Roberta Martin, Global Ecology, Carmegie Institution, Stanford, CA, Gregory P. Asner, Department of Global Ecology, Carnegie Institution for Science, Stanford, CA, Sandra Díaz, CONICET - Universidad Nacional de Córdoba, Instituto Multidisciplinario de Biología Vegetal, Córdoba, Argentina, Brian J. Enquist, Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ and Yadvinder Malhi, Environmental Change Institute, University of Oxford, Oxford, United Kingdom
Background/Question/Methods

The minor venation network of angiosperm leaves includes areoles (loops). These areoles may vary along trait axes describing loop size, shape, and topology. Variation in network architecture may influence leaf performance via functions related to network efficiency, redundancy, and cost.

We measured venation network trait space for 136 biomass-dominant angiosperm trees along a 3,300 m elevation gradient in eastern Peru using eight reticulation axes. We then examined the relative importance of multiple climatic, ecological, and evolutionary predictors of venation reticulation.

Results/Conclusions

Variation in minor venation network reticulation is constrained to a trait space with three axes that shows strong phylogenetic niche conservatism. These axes describe branching/reconnecting structures, elongated vs. compact areoles, and high vs. low density veins. Variation is best predicted by functional traits related to mechanical strength and secondary compound concentration, as well as by site temperature.

Angiosperm minor vein networks encompass a three-dimensional trait space that reflects tradeoffs in the geometry and prevalence of loops. Defensive, mechanical, and environmental factors primarily explain this variation, consistent with contemporary functions for these patterns. These results move beyond single-dimensional and single-function descriptions of venation networks.