To better understand palm hydraulic which traits are needed for palms to colonize warmer and drier biomes, we asked if there is a particular phenotype (as a combination of traits) that could be associated to each biome. For this, we took advantage of the unrivaled palm collection from the Royal Botanic Gardens Kew to sample palms from disparate biomes while controlling for phylogenetic relationships. For over 50 palm species in five continents, we collected data on at least eight functional traits (plant height and diameter, leaf area and leaf mass per area, stomata size and density, parallel and cross venation density). In addition to that, compiled and cleaned occurrence data for over 55K palms to calculate biome probability for over a thousand of palms species (~ 30% of all palm species). The first important result from PalmHydraulics project (Emilio et al, in prep) was to show that species that are able to colonize non-tropical rainforest biomes are equally numerous in the two main palm sub-families (Arecoideae and Coryphoideae). This was an unexpected result meaning that cold and dry affiliation on the palms family has evolved multiple times. Biome affiliation on palm species can be predicted from the traits palms possess. This is a very important result as by combining these models with species phenotypic plasticity is possible to identify the thresholds responses of palm species to climate and land cover changes.
To document the global patterns of palm abundance variation, we strength our collaboration with researchers from University of Aarhus Ecoinformatics and Biodiversity group (palm ecology and distribution experts) and University of Leeds Ecology and Global Change group (host of the largest tropical forest plot inventory repository). From the data collected by over a hundred field ecologists and botanists, we were able to assemble the most comprehensive (2,544 plots, 661,194 individual palms) dataset of arborescent palm abundance variation at the global scale. The second important result from PalmHydraulics project (Muscarella & Emilio et al. in preparation) was to show that palms are not abundant elsewhere as they are in the Neotropics. At least not large arborescent palms. Small understorey and climbing palms are largely unsampled in forests inventories and expect to increase the contribution of palms to biomass in over of 80% of the locations outside of the Neotropics. We also found that although it is not unexpected that estimates of aboveground biomass are more affected in locations were palm abundance is higher, the direction and strength of this effect are unpredictable, varying from +5% to -16%.
To gain a more mechanistic view on palm functioning we started a collaboration with a world-leading research group on plant hydraulics from INRA/University of Bordeaux (France). Together we performed in vivo visualization of embolism formation in palms for the first time. The main experiments we conducted at Soleil Synchrotron in the Spring of 2017, followed by additional experiments at BIOGECO in 2017 and 2018. The third important result (Emilio et al., under review) was to show that palms are not as vulnerable to drought-induced embolism (the main cause of plant mortality under drought) as we previously thought from their hydraulic structure. For the first time, we discover that palms show the entire range of embolism resistance of angiosperms in general. Besides that, we demonstrate with high-resolution micro-CT images that hydraulic functioning of palm leaves is supported by a large amount of water stored in their parenchyma.
