Climate and hydraulic traits interact to set thresholds for liana viability

Alyssa Willson and David Medvigy, Department of Biological Sciences, University of Notre Dame, Notre Dame, IN, Anna Trugman, Department of Geography, University of California Santa Barbara, Santa Barbara, CA, Jennifer S. Powers, Department of Ecology, Evolution, and Behavior, University of Minnesota, St. Paul, MN, Jennifer S. Powers, Department of Plant and Microbial Biology, University Of Minnesota, St. Paul, MN, Chris Smith-Martin, Ecology, Evolution and Environmental Biology, Columbia University, New York, NY

Background/Question/Methods
Trees and lianas dominate the canopy of tropical forests, competing for light and water. Their competition influences forest community structure and composition, with consequences for ecosystem services and forest management. These growth forms respond differently to variation in climate and resources. However, our understanding of the relevant mechanisms is limited and lianas have historically been left out of predictive ecosystem models. One factor limiting the inclusion of lianas in ecosystem models is the lack of compiled data on liana functional traits to understand how lianas differ from other plant functional types. We conducted a meta-analysis of liana functional traits, including traits related to leaves, stems, roots, and hydraulic architecture, that represent fundamental trade-offs in allocation and life history strategy. We compared liana and tree trait distributions to identify traits that differ between growth forms. We then developed a liana-tree competition model and parameterized the hydraulic traits using our meta-analysis. We used our model to simulate the hydraulic conductivity required to maintain positive annual net primary production (Kreq) for both lianas and trees under hydroclimate and competition scenarios representative of American tropical moist and tropical dry forests in present day and under projected end-of-century hydroclimate scenarios.
Results/Conclusions
We found that the observed difference between liana and tree hydraulic conductivities represents a critical source of inter-growth form variation in simulated net primary production. The simulated variation is more pronounced under drier atmospheric conditions and with larger canopy area to sapwood area ratios. Additionally, both lianas and trees are more sensitive to changes in vapor pressure deficit than soil water potential, suggesting that neither trees nor lianas are limited by soil water supply. Finally, under a future hydroclimate scenario assuming a 100% increase in vapor pressure deficit, we find that the simulated liana Kreq is greater than the estimated hydraulic conductivities of extant liana species, while tree Kreq remains nearly constant. This implies that lianas may reach a threshold for viability, with implications for tropical forest carbon cycling and ecosystem services. Liana hydraulic conductivity should be carefully considered in liana-enabled dynamic global vegetation models to accurately characterize tropical forest response to climate change.