Drought-related tree mortality is now a widespread phenomenon predicted to increase in magnitude with climate change. However, the patterns of which species and trees are most vulnerable to drought, and the underlying mechanisms have remained elusive, in part due to the lack of relevant data and difficulty of predicting the location of catastrophic drought years in advance. We used long-term demographic records and extensive databases of functional traits and distribution patterns to understand the responses of 20-53 species to an extreme drought in a seasonally dry tropical forest in Costa Rica, which occurred during the 2015 El Niño Southern Oscillation event. Overall, species-specific mortality rates during the drought ranged from 0% to 34%, and varied little as a function of tree size. By contrast, hydraulic safety margins correlated well with probability of mortality among species, while morphological or leaf economics spectrum traits did not. This firmly suggests hydraulic traits as targets for future research. K E Y W O R D S extreme drought, hydraulic traits, rainfall seasonality, tree mortality | 3123 POWERS Et al.

Summary There are two theories about how allocation of metabolic products occurs. The allometric biomass partitioning theory (APT) suggests that all plants follow common allometric scaling rules. The optimal partitioning theory (OPT) predicts that plants allocate more biomass to the organ capturing the most limiting resource. Whole‐plant harvests of mature and juvenile tropical deciduous trees, evergreen trees, and lianas and model simulations were used to address the following knowledge gaps: (1) Do mature lianas comply with the APT scaling laws or do they invest less biomass in stems compared to trees? (2) Do juveniles follow the same allocation patterns as mature individuals? (3) Is either leaf phenology or life form a predictor of rooting depth? It was found that: (1) mature lianas followed the same allometric scaling laws as trees; (2) juveniles and mature individuals do not follow the same allocation patterns; and (3) mature lianas had shallowest coarse roots and evergreen trees had the deepest. It was demonstrated that: (1) mature lianas invested proportionally similar biomass to stems as trees and not less, as expected; (2) lianas were not deeper‐rooted than trees as had been previously proposed; and (3) evergreen trees had the deepest roots, which is necessary to maintain canopy during simulated dry seasons.

Lianas are more abundant in seasonal forests than in wetter forests and are thought to perform better than trees when light is abundant and water is limited. We tested the hypothesis that lianas perform better than trees during seasonal drought using a common garden experiment with 12 taxonomically diverse species (six liana and six tree species) in 12 replicated plots. We irrigated six of the plots during the dry season for four years, while the remaining six control plots received only ambient rainfall. In year 5, we measured stem diameters for all individuals and harvested above‐ and belowground biomass for a subset of individuals to quantify absolute growth and biomass allocation to roots, stems, and leaves, as well as total root length and maximum rooting depth. We also measured rate of photosynthesis, intrinsic water use efficiency (iWUE), pre‐dawn and midday water potential, and a set of functional and hydraulic traits. During the peak of the dry season, lianas in control plots had 54% higher predawn leaf water potentials (ΨPD), and 45% higher photosynthetic rates than trees in control plots. By contrast, during the peak of the wet season, these physiological differences between lianas and trees become less pronounced and, in some cases, even disappeared. Trees had higher specific leaf area (SLA) than lianas; however, no other functional trait differed between growth forms. Trees responded to the irrigation treatment with 15% larger diameters and 119% greater biomass than trees in control plots. Liana growth, however, did not respond to irrigation; liana diameter and biomass were similar in control and irrigation plots, suggesting that lianas were far less limited by soil moisture than were trees. Contrary to previous hypotheses, lianas did not have deeper roots than trees; however, lianas had longer roots per stem diameter than did trees. Our results support the hypothesis that lianas perform better and experience less physiological stress than trees during seasonal drought, suggesting clear differences between growth forms in response to altered rainfall regimes. Ultimately, better dry‐season performance may explain why liana abundance peaks in seasonal forests compared to trees, which peak in abundance in less seasonal, wetter forests.

Recent findings suggest that tree mortality and post‐drought recovery of gas exchange can be predicted from loss of function within the water transport system. Understanding the susceptibility of plants to hydraulic damage requires knowledge about the vulnerability of different plant organs to stress‐induced hydraulic dysfunction. This is particularly important in the context of vulnerability segmentation between plant tissues which is believed to protect more energetically ‘costly’ tissues, such as woody stems, by sacrificing ‘cheaper’ leaves early under drought conditions. Differences in vulnerability segmentation between co‐occurring plant species could explain divergent behaviours during drought, yet there are few studies considering how this characteristic may vary within a plant community. Here we investigated community‐wide vulnerability segmentation by comparing leaf/shoot and stem vulnerability in all coexistent dominant canopy and understory woody species in a diverse dry sclerophyll woodland community, including multiple angiosperms and one gymnosperm. Previously published terminal leaf/shoot vulnerability to loss of water transport capacity was compared with stem xylem vulnerability to embolism measured on the same species at the same site. We calculated hydraulic safety margins for stems to determine variation in the risk of hydraulic failure during drought among species. The xylem of all species was found to be highly resistant to hydraulic dysfunction, with only two of the eight species exhibiting significantly different vulnerability to the overall mean. No evidence of vulnerability segmentation between shoots/leaves and stems was found in seven of the eight species. Phylogenetically diverse canopy and understory species in this evergreen sclerophyll woodland appear to have evolved similar strategies of drought resistance, including low xylem vulnerability to embolism and general lack of vulnerability segmentation. This convergence in hydraulic safety indicates a lack of hydraulic niche partitioning in this woodland community. A free plain language summary can be found within the Supporting Information of this article.

1. Despite their low contribution to forest carbon stocks, lianas (woody vines) play an important role in the carbon dynamics of tropical forests. As structural parasites, they hinder tree survival, growth and fecundity; hence, they negatively impact net ecosystem productivity and long-term carbon sequestration. 2. Competition (for water and light) drives various forest processes and depends on the local abundance of resources over time. However, evaluating the relative role of resource availability on the interactions between lianas and trees from empirical observations is particularly challenging. Previous approaches have used labour-intensive and ecosystem-scale manipulation experiments, which are infeasible in most situations. 3. We propose to circumvent this challenge by evaluating the uncertainty of water and light capture processes of a process-based vegetation model (ED2) including the liana growth form. We further developed the liana plant functional type in ED2 to mechanistically simulate water uptake and transport from roots to leaves, and start the model from prescribed initial conditions. We then used the PEcAn bioinformatics platform to constrain liana parameters and run uncertainty analyses. 4. Baseline runs successfully reproduced ecosystem gas exchange fluxes (gross primary productivity and latent heat) and forest structural features (leaf area index, aboveground biomass) in two sites (Barro Colorado Island, Panama and Paracou, French Guiana) characterized by different rainfall regimes and levels of liana abundance. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.